This Article Extract Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when Comments are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (12) Citing Articles via Google Scholar Google Scholar Articles by Zuckerman, S. Articles by LoSasso, A. Search for Related Content PubMed PubMed Citation Articles by Zuckerman, S. Articles by LoSasso, A. Related Collections Access To Care Hospitals

Health Tracking

TRENDS

How Did Safety-Net Hospitals Cope In The 1990s?

To survive these pressures, administrators of safety-net and other hospitals had to restructure to focus on new revenue opportunities, reduce some services, and become more efficient through mergers and streamlining of internal operations. Between 1989 and 1996, 190 hospital mergers occurred nationwide, involving approximately 450 hospitals.³ In addition, 60 percent of hospital administrators undertook one or more reengineering activities between 1990 and 1996, including reductions in personnel.⁴ Ultimately, if hospitals cannot improve their financial performance, they face the prospect of closing, as 295 hospitals did between 1990 and 1997.⁵

Safety-net hospitals often receive special subsidies through Medicare or Medicaid disproportionate-share hospital (DSH) funds, adjustments to payments for teaching, and tax appropriations or other grants. However, they may have fewer avenues to respond to financial pressures than non-safety-net hospitals have. In particular, they may be less likely to close because of their fundamental mission to serve the community. They also may be reluctant to drop services such as burn and trauma care, even though they do not generate revenues sufficient to cover costs, because there may be few alternative providers of these services or because they represent important components of an academic program. A further constraint for publicly owned safety-net hospitals may be strong unions that limit hospital administrators’ ability to streamline units and reduce staffing.⁶

A recent report by the Institute of Medicine (IOM) recognized threats to the continued viability of safety-net providers posed by the changing market and policy environment and called for research to monitor the effect of these changes.⁷ In this paper we describe some of the changes safety-net hospitals made in the face of industry wide pressures during 1990–1997, and the financial consequences. We compare safety-net hospitals with other hospitals to assess how the special role of safety-net hospitals may affect their ability or willingness to respond strategically.

Identification of safety-net hospitals. The first step in this study was to identify safety-net hospitals. The IOM study committee used a general approach, defining the safety net as "providers that organize and deliver a significant level of health care and other related services to uninsured, Medicaid and other vulnerable patients."⁸ However, conducting statistical analyses using this general definition is not straightforward.

Some researchers have identified certain categories of hospitals as safety-net based on their ownership, teaching status, or geographic location. For example, public hospitals and urban academic medical centers are often labeled as safety-net institutions.⁹ Clearly, large public teaching hospitals are likely to be important providers of uncompensated care in many urban areas.¹⁰ However, using organizational characteristics such as "public" or "teaching" may be misleading. For example, in markets with limited public hospital capacity or without major teaching hospitals, nonpublic and nonteaching hospitals may be major providers of uncompensated care; also, some public or major teaching hospitals may be located in wellinsured communities and may thus provide minimal uncompensated care. Recognizing this problem, researchers have begun to look at the actual pattern and level of uncompensated care delivered by particular hospitals.

Some researchers have identified safety-net hospitals based on the proportion of a hospital’s total resources devoted to Medicaid enrollees or patients for whom the hospital was not compensated. Linda Fishman and James Bentley referred to this as "hospital burden."¹¹ Other researchers have focused on how dependent the community is on a particular hospital for treating its poor population. ¹² Together, these two concepts provide different perspectives (hospital and community) on indigent care that are related to the IOM definition.

Data sources and methods. Because both the hospital and community perspectives are important, this paper melds the concepts into a unified approach for identifying groups of urban safety-net hospitals. Our primary data source for the identification process was the 1990 American Hospital Association (AHA) Annual Survey of Hospitals.¹³ Our sample included all urban, nonfederal, short-term general hospitals (2,668 hospitals). We defined uncompensated care as the sum of charity care and bad-debt expense.¹⁴ Charity care and bad debt were reported as gross revenues or charges on the AHA annual survey. We adjusted uncompensated care charges by a hospital-specific ratio of cost to charges to compute uncompensated care expense.¹⁵

Our market share measure represents an "adjusted" measure because a hospital’s expected market share depends on the number of hospitals in its metropolitan statistical area (MSA). We neutralize this effect by multiplying a hospital’s uncompensated care market share by the number of hospitals in its market. If uncompensated care were equally distributed across all hospitals, these adjusted market shares would all equal 1. Hospitals with high uncompensated care market share are those with an adjusted market share of at least 2, approximately the eighty-fifth percentile of this measure for hospitals under study. Hospital burden is the percentage of hospitals’ total expenses that are uncompensated.¹⁶ As did Fishman and Bentley, we use the top decile of the distribution of hospital burden nationwide as a threshold for this dimension.¹⁷

Our method produces four groups of hospitals, which form the basis for our study.¹⁸ These four groups are (1) Group 1, high market share/high burden: The 107 safety-net hospitals that met both criteria in 1990 are unique in terms of the large community role they play in providing uncompensated care and in the large institutional commitment they have made to indigent care. (2) Group 2, high market share only: In 1990, 231 hospitals met only the high market share criteria but not the uncompensated care burden threshold. These play a major role in their community in uncompensated care but have paying patients who generate revenues that cover a large proportion of their expenses. (3) Group 3, high burden only: The 160 hospitals in 1990 that had high uncompensated care burden but not high market share may serve as safety nets for their immediate communities but do not treat a large share of indigent patients in the broader market. (4)Group 4, non-safety-net hospitals: The remaining 2,170 hospitals were by definition non-safety-net hospitals in 1990.

While this approach has advantages over using organizational labels, it also has limitations. By looking at uncompensated care provision and patterns in the 1990 base period, we do not explicitly recognize that there is year-to-year variation in an institution’s commitment to and involvement in uncompensated care. However, we identify a set of institutions that were important providers for persons who were unable to pay for care during the base period. In addition, the comparisons we make between the safety-net and non-safety-net groups do not attempt to control for differences in market characteristics that could affect behavior and performance. For example, safety-net hospitals could be less likely than non-safety-net hospitals to drop certain services because demand for these services is greater in the markets in which the safety-net hospitals are located.

Characteristics Of Safety-Net Hospitals

 
Both the high market share/high burden (Group 1) and the high market share only (Group 2) safety-net hospitals are much larger and more likely to be members of the Council of Teaching Hospitals (COTH) or have residency programs than are the other two groups of hospitals (Exhibit 1). From an ownership perspective, 70 percent of Group 1 hospitals are public hospitals and none are for-profit institutions, while almost 85 percent of Group 2 hospitals are private, not-for profit. About one-third of high burden only (Group 3) safety-net hospitals are public hospitals —three times the rate of non-safety-net hospitals (Group 4)—but these smaller hospitals are not significantly different from non-safety- net hospitals with respect to teaching status.

Uncompensated care burden. Patterns of uncompensated care provision for various safety-net hospital categories shown in Exhibit 1 reflect the definitions for each group. Group 1 hospitals were the largest providers of uncompensated care, providing an average of $25.1 million in uncompensated care in 1990. Group 2 hospitals provided an average of $8.9 million—almost 30 percent of the uncompensated care in their markets. Group 3 hospitals provided $4.0 million in 1990, twice the average of Group 4 hospitals but well below the levels of the other groups.

Specialty services. Provision of selected specialty services also varied across hospital groups (Exhibit 1).With very limited exceptions, the two groups of hospitals with high uncompensated care market shares were significantly more likely to provide all of these services in 1990. Interestingly, Group 3 hospitals were less likely to provide maternity care or have a trauma center, acquired immunodeficiency syndrome (AIDS) services, an emergency room, or a neonatal intensive care unit (ICU) compared with Group 4 hospitals. This suggests that hospitals can be burdened with provision of uncompensated care even though they do not offer many of the services often provided by safety-net hospitals.¹⁹ Also, not all of the specialty services are provided by safety-net hospitals. For example, 34 percent of burn units and 62 percent of trauma centers are at non-safety-net hospitals, because these hospitals greatly outnumber safety-net hospitals.

Organizational And Operational Changes

Top Characteristics Of Safety-Net... Organizational And Operational... Discussion NOTES

 
Closures and mergers. Overall, 154 urban hospitals (5.7 percent) closed between 1990 and 1997, and 321 (12 percent) merged with other facilities (Exhibit 2).²⁰ Among the 338 safety-net hospitals in Groups 1 and 2, only two closed during the study period, but fifty-nine were involved in mergers. The merger rate for Group 2 hospitals significantly exceeded that for non-safety-net hospitals. Of the 160 Group 3 hospitals, fourteen closed, and only ten merged by 1997. The merger rate for these hospitals was significantly lower than that for non-safety-net hospitals.

Operational and financial changes. Group 1 hospitals differed significantly from non-safety-net hospitals along several operating and financial dimensions (Exhibit 4). Perhaps most noteworthy, for hospitals that provided maternity care in both 1990 and 1997, the average number of births at Group 1 hospitals fell by 4.8 percent annually but remained stable at non-safety-net hospitals. This loss of births could have limited Group 1 hospitals’ access to a potential pool of recently eligible Medicaid patients.²³ It is also consistent with the failure of hospitals in this group to raise their Medicaid share of total days compared with that for non-safety-net hospitals. Group 1 hospitals also had slower growth in outpatient visits, a reduction in emergency room visits, and a decrease in Medicare share relative to non-safety-net hospitals. Other volume measures—admissions and inpatient days—changed at virtually the same rate across all groups.

Group 2 hospitals are statistically similar to non-safety-net hospitals in terms of changes in service volume, expenses, and net patient care revenues. The only exception is that Group 2 hospitals had a significant decline in births relative to non-safety-net hospitals. Both experienced improvements in total margin, but the improvement was significantly smaller for Group 2 hospitals.²⁴ These hospitals also had slower growth in uncompensated care expenses and reduced uncompensated care market share relative to the non-safety-net hospital group. However, because their 1990 base was higher, the absolute average amount of uncompensated care that each Group 2 hospital provided in 1997was still greater than among non-safety-net hospitals.

The Group 3 hospitals that continued operations through the 1990s did not experience service volume changes that were significantly different from those in non-safety-net hospitals (Exhibit 4). However, these hospitals did curtail the growth in their uncompensated care expenses relative to those of non-safety- net hospitals. Also, Group 3 hospitals’ expense and revenue growth rates over the period were significantly greater than those of Group 4 hospitals. The slower growth in uncompensated care expenses combined with the more rapid growth in total expenses led to a significant reduction in uncompensated care burden.²⁵ Nonetheless, total margins fell for Group 3 hospitals relative to non-safety-net hospitals.²⁶

Discussion

Top Characteristics Of Safety-Net... Organizational And Operational... Discussion NOTES

 
Safety-net hospitals are heterogeneous and undergo different organizational, operational, and financial changes, which have implications for their future viability. Hospitals with high market share/high burden remained the most important providers of uncompensated care and, despite stagnant growth in admissions and losses in the number of births relative to non-safety-net hospitals, virtually never closed. Hospitals with high market share only continued their important role but did reduce their share of uncompensated care relative to non-safety-net hospitals. These facilities also appeared to be the most attractive merger partners, which suggests that being a large provider of indigent care in a market area is not a barrier to merger. Hospitals with high burden only were most at risk of closing. Even among those that remained open, despite curtailing their uncompensated care burden and achieving significantly greater growth in patient revenues, margins fell relative to those of non-safety-net hospitals.

Implications of service changes. Our finding that safety-net hospitals reduced the provision of some services during the study period may reflect internal reorganization or consolidation of services with affiliated hospitals. ²⁷ These service changes also may reflect changes in technology or treatment modality that allowed care to shift from a hospital to another setting. To the extent that these explain the observed patterns, then the hospitals’ management response should be seen in a positive light. However, hospitals may be dropping some of these specialty services to reduce their exposure to uninsured patients, and this could curtail access.

The changes we observed in maternity care could be ominous for safety-net hospitals, especially those in the groups with high uncompensated care market share. At a time when Medicaid was paying for an increasing share of births, these hospitals were providing significantly fewer births. In contrast, non-safety- net hospitals that provided maternity care in 1990 had no reduction in births, and a significant share added maternity care. To the extent that Medicaid payments for births are greater than the marginal costs of providing the service, safety-net hospitals are losing revenues that could be used to cross-subsidize uncompensated care. However, prior research suggests that these effects might be small.²⁸

Medicaid DSH payments. It appears that Medicaid DSH payments may have allowed hospitals with high burden/high market share to survive in the face of declining patient volumes, although with deteriorated financial positions relative to those of non-safety- net hospitals.²⁹ Federal DSH payments to states increased from $801 million in 1990 to $8.9 billion in 1997.³⁰ Although not all of the funds were distributed to hospitals, the expansion in many states increased total Medicaid hospital payments.³¹ Among hospitals, those with high market share/high burden would be most likely to receive these supplements.

Hospitals with high burden/high market share could have been adversely affected if Medicaid DSH policy changes in the 1997 BBA had been kept in place. Under those rules, Medicaid DSH payments were to have been constrained, with expected reductions of 20 percent between 1998 and 2002.³² Congress passed the Benefits Improvement and Protection Act (BIPA) of 2000 and eliminated the biggest Medicaid DSH cuts that were planned for 2001 and 2002. In fact, BIPA raised Medicaid DSH allotments in states with very low DSH spending and gave states the ability to increase Medicaid DSH payments to public hospitals.³³ These rollbacks suggest that Congress recognized the role these payments play in financing public safety-net hospitals.

Impact of managed care. The period we studied represented a time of substantial growth in Medicaid health maintenance organization (HMO) enrollment—Medicaid HMO penetration increased from 12 percent in 1990 to 39 percent in 1997.³⁴ The introduction of managed care into Medicaid has had a negative effect on safety-net hospitals, by redistributing Medicaid patients to non-safety-net providers and by reducing payments to safety-net providers through negotiated rates.³⁵ Our findings suggest that the Medicaid managed care expansion may not have had a heavy financial toll on major safety-net hospitals between 1990 and 1997 even though the volume of several hospital services fell for some hospitals. Since 1997, however, some states have expanded enrollment to include different populations (for example, the disabled), while others have continued implementation of mandatory HMO programs. Thus, the threats to safety-net hospitals’ viability associated with Medicaid managed care are likely to increase over time.

Staffing levels and expenses. Our findings suggest that although hospitals with high market share/high burden lost service volume, they did not reduce staffing levels and expenses commensurately. This may represent a decrease in efficiency resulting from a lack of management flexibility. However, it also could indicate that these facilities maintain excess capacity to be able to respond to uncertain demand or that they treat a greater number of sicker patients. One limitation of our study is that we did not examine changes in efficiency of hospital operations over time. Evidence from the late 1980s suggests that hospitals with low margins respond to payment pressures by increasing efficiency.³⁶ Thus, reductions in payment generosity through limits on DSH spending and growth in managed care may force higher levels of efficiency in safety-net hospitals, rather than causing a reduction in essential services to needy populations. If safety-net hospitals can respond by improving efficiency, their viability and patient care may be less affected by environmental changes, at least for small levels of change. This along with continued careful tracking and monitoring of the U.S. hospital safety net are important areas for future study.

Editor's Notes

 
Stephen Zuckerman is principal research associate at the Urban Institute’s Health Policy Center in Washington, D.C. Gloria Bazzoli is a professor of health administration, Medical College of Virginia/Virginia Commonwealth University. Amy Davidoff is a research associate at Urban. Anthony LoSasso is research assistant professor, Institute for Health Services Research and Policy Studies, Northwestern University.

This paper, as part of the Urban Institute’s Assessing the New Federalism project, has received funding from the Annie E. Casey, W.K. Kellogg, Robert Wood Johnson, Henry J. Kaiser Family, Ford, John D. and Catherine T. MacArthur, Charles Stewart Mott, and David and Lucile Packard Foundations and several other foundations. This research was conducted in collaboration with the Health Research and Educational Trust (HRET) of the American Hospital Association, which undertook aspects of the data analysis. The authors are particularly grateful for the programming support provided by Ray Kang of HRET. This research was conducted while Gloria Bazzoli was research professor with the Institute for Health Services Research and Policy Studies at Northwestern University. The opinions expressed are those of the authors and do not represent the views of the Urban Institute, Northwestern University, Virginia Commonwealth University, the American Hospital Association, their trustees, or their sponsors.

NOTES

Top Characteristics Of Safety-Net... Organizational And Operational... Discussion NOTES

 

1. G. Melnick, J. Mann, and A. Bamezai, "Preliminary Findings: Hospital Uncompensated Care Not Keeping Pace with Per Capita Spending," HCFO News and Progress (March 2000).
2. R.J. Baxter and R.E. Mechanic, "The Status of Local Health Care Safety Nets," Health Affairs (July/Aug 1997): 7–23.
3. G. Bazzoli et al., "Hospital Reorganization and Restructuring Achieved through Merger," Institute for Health Services Research and Policy Studies Working Paper no. 00-09 (Evanston, Ill.: Northwestern University, 2000).
4. S. Walston, L. Burns, and J. Kimberley, "Does Reengineering Really Work? An Examination of the Context and Outcomes of Hospital Reengineering Initiatives," Health Services Research (February 2000): 1363–1388.
5. Bazzoli et al., "Hospital Reorganization."
6. Baxter and Mechanic, "The Status of Local Health Care Safety Nets."
7. Institute of Medicine, America’s Health Care Safety Net: Intact but Endangered (Washington: National Academy Press, 2000).
8. Ibid.
9. Baxter and Mechanic, "The Status of Local Health Care Safety Nets"; and L.E. Fishman and J.D. Bentley, "The Evolution of Support for Safety-Net Hospitals," Health Affairs (July/Aug 1997): 30–47.
10. J.M. Mann et al., "A Profile of Uncompensated Hospital Care, 1983–1995," Health Affairs (July/Aug 1997): 223–232.
11. Fishman and Bentley, "Evolution of Support."
12. Ibid.; P.J. Cunningham and H.T. Tu, "A Changing Picture of Uncompensated Care," Health Affairs (July/Aug 1997): 167–175; and G. Atkinson, D. Helms, and J. Needleman, "State Trends in Hospital Uncompensated Care," Health Affairs (July/Aug 1997): 233–241.
13. American Hospital Association, Annual Survey of Hospitals (Chicago: AHA, 1990 and 1997).
14. Thomas Rundall and colleagues found that variation in accounting practices across hospitals made it impossible to make a meaningful distinction between charity care and bad debt. T. Rundall, S. Sofaer, and W. Lambert, "Uncompensated Hospital Care in California: Private and Public Hospital Responses to Competitive Market Forces," Advances in Health Economics and Health Services Research 9 (1988): 113–133.
15. One difficulty with the AHA annual survey is the underreporting of confidential financial variables such as the components of uncompensated care. About 15 percent of urban, nonfederal, short-term general hospitals did not provide data that allowed us to compute uncompensated care in 1990. To derive a value of uncompensated care for hospitals with missing values, we used a regression-based prediction. The model—based on previous research by the authors—included as independent variables market demographics, hospital market characteristics, hospital characteristics, and a series of state indicator variables. The regressions were estimated separately for three ownership groups, using ordinary least squares, with R-squares for the regressions ranging from 0.60 to 0.85. This process was repeated for 1997. Details are available from Stephen Zuckerman, <szuckerm{at}ui.urban.org>.
16. The IOM definition of safety-net providers would have suggested including both Medicaid and other vulnerable populations as well as uncompensated care patients in our definition. We focus on uncompensated care provision as opposed to Medicaid or care to specific vulnerable groups. In part, our rationale is that Medicaid patients have insurance coverage. Although base Medicaid payment rates have historically been low relative to those of private payers, they tend to cover a high percentage of patient costs. Medicare Payment Advisory Commission, Report to the Congress: Selected Medicare Issues (Washington: MedPAC, June 2000), 186.
17. In 1990 hospitals with an uncompensated care burden of 9.2 percent of expenses were included in the high-burden group.
18. Examples of hospitals falling into our three categories of safety-net hospitals include the following (from case studies conducted as part of the Urban Institute’s Assessing the New Federalism study): Boston City Hospital (now Boston Medical Center) falls into the high market share/high burden group (Group 1). This large, formerly public hospital has historically played an important role in indigent care delivery. Massachusetts General Hospital falls into the high market share only group (Group 2). This is a large teaching hospital in Boston that because of its size provides a substantial amount of uncompensated care as well as a substantial amount of care to insured populations. Somerville Hospital (now part of the Cambridge Health Alliance) falls into the high burden only group (Group 3). This small, nonprofit hospital is not a major teaching institution but provides a disproportionately large share of the Medicaid inpatient days in the Boston MSA relative to its size. See N. Brennan, S. Guterman, and S. Zuckerman, How Are Safety Net Hospitals Responding to Health Care Financing Changes? (Washington: Urban Institute, 2001).
19. D. Gaskin, Safety Net Hospitals: Essential Providers of Public Health and Specialty Services (New York: Commonwealth Fund, February 1999).
20. Hospital closures include the small number of facilities that converted to non–acute care services, effectively leaving the hospital market.
21. The share of Group 3 hospitals providing maternity care actually increased by the greatest magnitude (6.2 percentage points), but the change was not quite statistically significant (p > .05).
22. As a result of expansions in Medicaid eligibility, the proportion of births financed by Medicaid rose from 15 percent in 1985 to 32 percent in 1991 and to 39 percent in 1995. As a result of welfare reform and the associated drop in Medicaid enrollment, only 35 percent of births were covered by Medicaid in 1997. See J. Frost et al., State Implementation of the Medicaid Eligibility Expansions for Pregnant Women (New York: Alan Guttmacher Institute, 1993); National Governors’ Association, States Have Expanded Eligibility through Medicaid and the State Children’s Health Insurance Program (Washington: NGA, 1999); and NGA, Income Eligibility for Pregnant Women and Children (Washington: NGA, 2000).
23. Other studies have also documented these types of patterns. See L. Fagnani and J. Tolbert, The Dependence of Safety Net Hospitals and Health Systems on the Medicare and Medicaid Disproportionate Share Hospital Payment Programs, Pub. no. 351 (New York: Commonwealth Fund, October 1999); and D. Gaskin, J. Hadley, and V. Freeman, "Are Urban Safety-Net Hospitals Losing the Competition for Low-Risk Medicaid Patients?" Institute for Health Care Research and Policy Working Paper no. 98-107 (Washington: IHCRP, October 1999). In fact, Gaskin and colleagues show that the Medicaid maternity patients that safety-net hospitals are losing tend to be low-risk maternity patients, leaving safety-net hospitals with an increased share of high-risk patients.
24. In fact, the increase in total margins for Group 2 hospitals was not statistically significant.
25. To some extent, this reduction could be due to regression to the mean. This is possible because these hospitals were included in the group as a result of a high uncompensated care burden in a single base year. However, these hospitals had much slower growth in nonoperating revenues (data not shown), and, as result, they experienced no change in total margin.
26. It is worth noting why the financial data in Exhibit 3 indicate the seemingly contradictory results that net patient revenue grew at a faster rate than total expenses, yet average total margins fell over the same period. The revenue and expense figures did not start at the same level in 1990, so the percentage changes are not directly comparable. In addition, because these are expressed as average percent changes, they should not translate directly into the absolute change in total margins. Our choice to present the expense and revenue figures as we have was based on our desire to represent the circumstances of the average hospital within each safety-net group, as opposed to the aggregate experience of each group.
27. G. Bazzoli et al., "A Taxonomy of Healthcare Networks and Systems: Bringing Order Out of Chaos," Health Services Research 33, no. 6 (1999): 1683–1717.
28. A. Davidoff et al., "The Effect of Changing State Health Policy on Hospital Uncompensated Care," Inquiry (Fall 2000): 253–267.
29. Medicare DSH payments also increased, but the relative expenditures on Medicaid DSH are much larger than those on Medicare DSH.
30. T. Coughlin, L. Ku, and J. Kim, Reforming the Medicaid Disproportionate Share Hospital Program in the 1990s, Assessing the New Federalism Discussion Paper no. 26 (Washington:Urban Institute, January 2000).
31. T. Coughlin and L. Ku, "Medicaid Special Disproportionate Share and Other Financing Programs," Health Care Financing Review (Spring 1995): 27–54.
32. T. Coughlin and D. Liska, "State Responses to Changes in Federal Medicaid DSH Policies: Highlights from ANF Study States" (Presented at the Urban Institute and Health Affairs Assessing the New Federalism Conference, Washington, D.C., June 1998).
33. As our analysis suggests, not all hospitals with high market share/high burden are public hospitals. However, BIPA allowed states to increase gross Medicaid DSH payments from 100 percent of uncompensated costs (including shortfalls related to low Medicaid payment rates) to 175 percent for public hospitals, which represent a large share of this safety-net hospital category. Because Medicaid DSH payments are often financed by intergovernmental transfers (IGTs) from public hospitals receiving the payments, and because not all federal payments are received by the hospitals providing the IGT, the net Medicaid DSH payments received by public hospitals may be well below 50 percent of their gross DSH payments.
34. Calculation based on information available at the Health Care Financing Administration (HCFA) Web site, <www.hcfa.gov/medicaid/omchmpg.htm>.
35. Gaskin et al., "Are Urban Safety-Net Hospitals Losing the Competition for Low-Risk Medicaid Patients?"
36. J. Hadley, S. Zuckerman, and L. Iezzoni, "Financial Pressure and Competition: Changes in Hospital Efficiency and Cost-Shifting Behavior," Medical Care (March 1996): 205–219.

                      What's this?

This article has been cited by other articles:

				J. S. Ross, S. S. Cha, A. J. Epstein, Y. Wang, E. H. Bradley, J. Herrin, J. H. Lichtman, S.-L. T. Normand, F. A. Masoudi, and H. M. Krumholz Quality Of Care For Acute Myocardial Infarction At Urban Safety-Net Hospitals Health Aff., January 1, 2007; 26(1): 238 - 248. [Abstract] [Full Text] [PDF]

				G. J. Bazzoli, R. Kang, R. Hasnain-Wynia, and R. C. Lindrooth An Update On Safety-Net Hospitals: Coping With The Late 1990s And Early 2000s Health Aff., July 1, 2005; 24(4): 1047 - 1056. [Abstract] [Full Text] [PDF]

				C. C. Havighurst Starr on the Corporatization and Commodification of Health Care: The Sequel Journal of Health Politics Policy and Law, August 1, 2004; 29(4-5): 947 - 968. [PDF]