Reforming the Medicaid Disproportionate Share Hospital Program

Abstract

Since 1991, three Federal laws have sought to reform the Medicaid disproportionate share hospital (DSH) program, which is designed to help safety net hospitals. This article provides findings from a 40-State survey about Medicaid DSH and supplemental payment programs in 1997. Results indicate that the overall size of the DSH program did not grow from 1993 to 1997, but the composition of DSH revenues and expenditures changed substantially: A much higher share of the DSH funds were being paid to local hospitals and relatively less was being retained by the States. The study also revealed that large differences in States' use of DSH still persist.

Introduction

The Medicaid DSH program is designed to help safety net hospitals that serve large numbers of Medicaid and uninsured patients. In 1997, States and the Federal Government spent $15.9 billion on Medicaid DSH payments, accounting for nearly 10 percent of total Medicaid spending. For the past decade the DSH program has been a subject of considerable controversy, sometimes sparking intense debate between States and the Federal Government. Indeed, on three separate occasions since 1991, Congress has passed legislation expressly aimed at reforming the Medicaid DSH program. Most recently, DSH was a key issue in the 1997 Federal budget discussions and major changes were made to the program, including Federal cutbacks.

Federal policymakers had several concerns with the DSH program. Perhaps most prominent was that monies being paid out through the DSH program were not always being used to help safety net hospitals; instead, States were often keeping Federal DSH funds to be used for other purposes. Indeed, a 1993 survey of 39 State DSH programs showed that at least one-third of Federal DSH funds were retained by States (Ku and Coughlin, 1995).

Despite the fact that the DSH program has been a controversial policy topic, only limited information on the program is at present available.¹ Updated information about States' DSH programs—such as how States are financing their programs or what types of hospitals are receiving DSH funds—are not readily available. Further, little is known about how States intend to handle the cuts in Federal DSH payments that were included as part the 1997 Balanced Budget Act (BBA). Between 1998 and 2002, Federal Medicaid DSH payments will be reduced by more than $10 billion. How States deal with the BBA DSH cuts have important implications for the financial health of many safety net hospitals nationwide as these providers rely on Medicaid DSH funds to support their operations, including rendering care to the uninsured.

In an effort to fill this information gap we conducted a survey of State Medicaid DSH programs. A better understanding of the current structure of States' DSH programs would be most helpful to policymakers in the design and targeting of any future reforms of the DSH program. Building on a study of States' DSH programs conducted in 1994 (Ku and Coughlin, 1995), we sent a survey to Medicaid agencies to all 50 States and the District of Columbia in spring 1998. Among other things, the survey asked how States fund their DSH programs and what types of hospitals receive DSH payments. The survey also asked about other Medicaid provider supplemental payments (beyond DSH) that States may make. In addition, this study entailed conducting phone interviews with State Medicaid officials in seven States to find out further details about their DSH programs and policies.

Before presenting our survey findings, we present a brief background on the Medicaid DSH program and a description of our study methods. Then we present our results in five sections: The first three discuss how States financed their DSH programs and distributed DSH payments among different types of hospitals. The third section examines how DSH expenditures are allocated across States. The fourth section compares our findings to those from a previously reported survey of States' DSH programs for 1993. The final results section describes our findings on Medicaid provider supplemental programs. The article concludes with a discussion of the policy implications of the study results.

Background

Under Medicaid law States are required to take into account the situation of hospitals which serve a disproportionate number of low-income patients with special needs when setting inpatient hospital payment rates. The rationale behind the special payments is that hospitals rendering high volumes of care to Medicaid recipients often lost money because of historically low Medicaid reimbursement rates. They also lost money because these hospitals are often the same facilities that provide high volumes of care to indigent patients, causing them to have high levels of uncompensated care. In the early 1980s, Congress established the Medicaid DSH program to provide some financial relief to hospitals serving the poor. Another goal of the DSH program was to maintain hospital access for the poor: By helping support hospitals that serve large numbers of the poor, it was hoped these hospitals could continue to operate, and access for the indigent would not deteriorate.²

Like many other features of the Medicaid program, States are allowed considerable freedom in designing their DSH program. For example, they are given a great deal of latitude in determining which hospitals qualify for DSH payments. They also have substantial discretion in setting the level of DSH payments made to individual hospitals. At a minimum, though, Federal law mandates that States have a DSH program and must make payments to hospitals that have a Medicaid inpatient use rate of at least one standard deviation above the mean for the State or a low-income inpatient use rate of 25 percent or more. However, States can go beyond the Federal minimum criteria and make DSH payments to hospitals with Medicaid inpatient use rates as low as 1 percent. Because of this flexibility, States' DSH programs vary greatly both in how DSH payments are rendered and the types of hospitals that receive payments.

State DSH programs also vary because beginning in the late 1980s many States started to use novel financing mechanisms—such as provider taxes and donations, and, later, intergovernmental transfers (IGTs) and certified public expenditures (CPEs)—to help finance their DSH programs.³ Under these mechanisms revenues obtained from hospitals in the form of provider taxes and donations, IGTs or CPEs were used as the State's share of its DSH payments. For example, a typical transaction could work as follows: A State receives $10 million in revenue—in the form of a tax, IGT or CPE—from a hospital. The State then makes a $12-million DSH payment back to the provider. Assuming the State has a 50-percent Federal matching rate, the State would get $6 million in Federal Medicaid funds. At the end of the transaction, the provider has netted $2 million dollars ($12 minus $10) in DSH payments, all from Federal funds. The State has received $4 million in Federal money without spending any of its own funds. The Federal Government has paid $6 million in DSH payments, but only $2 million was actually gained by the hospital.

It should be noted that use of such mechanisms is largely restricted to financing the DSH program. Generally, States finance their share of other parts of the Medicaid program (such as inpatient hospital care, physician care, etc.) with monies from State general revenue. While several States require local government participation in supporting the Medicaid program, local financing has not historically relied on financing mechanisms used to fund DSH. In 1994, about 22 States required county participation to fund the non-DSH part of Medicaid. Among those, only New Mexico's counties used IGTs to fund its county Medicaid share (U.S. Advisory Commission on Intergovernmental Relations, 1994).

The ability to leverage Federal dollars through the DSH program prompted many States to establish large programs which relied on creative financing in the early 1990s. Reflecting this trend, DSH payments went from $1.4 billion in 1990 to $17.5 billion in 1992. To slow the growth of DSH payments, Congress passed legislation in 1991 and 1993. These laws essentially banned provider donations and capped States' use of provider taxes. The law also required that provider tax programs be a real tax. That is, taxes had to be “broad-based” across all providers in a health care class, and DSH payments could not be constructed to hold the provider “harmless” for the cost of the tax. In addition, Congress capped the maximum growth in Federal DSH payments made to States. Finally, the laws limited how much an individual hospital can receive in DSH payments to no more than 100 percent of a hospitals's unreimbursed costs of providing care to Medicaid patients and uninsured. This limit became known as the hospital-specific cap and was in full effect by 1995.

States used the DSH program for a range of purposes, including financing other health programs and general State fund expenditures. One area where several States made significant use of the DSH program was to help finance uncompensated care costs for uninsured patients in institutions for mental disease (IMDs), particularly State-owned facilities. The U.S. General Accounting Office, for example, estimated that between 1993 and 1995, Medicaid DSH payments to IMDs increased by about $1 billion (U.S. General Accounting Office, 1997). Congress was particularly upset with IMD DSH payments as Medicaid historically has played a limited role in financing care provided in mental hospitals: By law, Medicaid does not pay for inpatient psychiatric care of adults between the ages of 21 and 64. Instead, patient care in IMDs has traditionally been a State or local responsibility. To curb States' use of the Medicaid DSH program to fund IMDs, Congress included provisions in the 1997 BBA limiting how much DSH money IMDs can receive. By 2002, no more than 33 percent of a State's Federal DSH allotment can be paid to these facilities. For some States, such as Kansas where IMD DSH payments accounted for 90 percent of its total DSH payments in 1997, the IMD provision will require a major restructuring of their DSH programs.

The BBA contains several other provisions which will affect States' DSH programs in the near future. Most prominently, as previously mentioned, the BBA called for Federal reductions in DSH payments. For years 1998-2002, BBA set out State-specific Federal DSH allotments. According to the U.S. Congressional Budget Office, these new Federal allotments represent a $10.4-billion reduction in Federal spending during the period 1998-2002 (U.S. Congressional Budget Office, 1997). The $10.4 billion figure assumed that 25 percent of Federal DSH savings as specified in the new provisions would not be realized because States would make up some of the DSH savings by spending more in other parts of their Medicaid programs (U.S. Congressional Budget Office, 1997). The cuts were not evenly distributed across the States. For example, one simulation, showed that, on average, States' DSH spending in 2002 will be 19 percent lower than what it was in 1995. However, the percentage reductions ranged from zero for some States—such as Delaware, Minnesota, and Wisconsin—to more than 20 percent for others—such as Colorado, Connecticut, and Kansas (Coughlin and Liska, 1997). In general, cutbacks were associated with the size of a State's DSH program: the bigger the program the bigger the cutbacks.

The BBA also contained longer term DSH spending limits. After 2002, the BBA allows Federal DSH expenditures to increase by the percentage change in the Consumer Price Index, subject to a ceiling of 12 percent of the Federal share of each State's total annual Medicaid expenditures. Under previous law, States that had fallen under the 12 percent ceiling were allowed to increase DSH spending at the same rate as their overall Medicaid expenditure growth. The U.S. Congressional Budget Office estimated that the BBA DSH limits will save the Federal Government about $40 billion over the next 10 years.

While some States have used DSH for purposes other than helping support hospitals, Medicaid DSH payments still represent an important revenue source for safety net hospitals. A recent survey, for example, showed that without Medicaid DSH payments, public hospitals would have experienced a 13-percent loss on Medicaid payments in 1996 (Fagnani and Tolbert, 1999). With Medicaid DSH payments, the surveyed public hospitals reported a positive Medicaid margin of 6 percent. Thus, DSH payments not only made up for the shortfall in Medicaid reimbursement but also provided hospitals with extra funds to help pay for other costs such as uncompensated care.

In recent years, some States have developed “supplemental payment programs” that are akin to DSH, but not subject to the same legislative limitations (Coughlin and Liska, 1998). These are payments, made in addition to regular Medicaid reimbursements, to certain types of health care institutions, including hospitals, nursing facilities or intermediate care facilities for the mentally retarded. For example, a State might use supplemental payments to cover the gap between usual Medicaid reimbursement and the Medicare upper payment limit.⁴ Like DSH, in many cases, providers that receive supplemental payments may contribute revenue to the State Medicaid agency to help cover the State share of the payments. Given the pending Federal caps on DSH funds under the BBA, it is important to understand the new alternatives to DSH that have been developed by some States.

Methods

To conduct this study we sent a survey to all State Medicaid agencies asking them to describe their DSH programs. Among other things, we asked States to list the types of revenue sources they used to finance their DSH programs in State fiscal year 1997. Revenue sources included State general funds, health care provider taxes, IGTs or CPEs from county or local entities, and fund transfers or CPEs from State entities. We also asked how much revenue they received from each funding source.

In addition to funding sources, we sought information about the level of State fiscal year 1997 DSH payments made to various types of inpatient hospitals, including private acute hospitals, local or county acute hospitals, State acute hospitals, private mental hospitals, local or county mental hospitals, and State mental health hospitals.

Beyond the DSH program the survey queried States whether they made any other supplemental payments (that are not reported as DSH payments) to providers over and above regular Medicaid reimbursement. We sought information on supplemental payment programs for all types of Medicaid services and providers. If a State indicated they had such a program, we asked them to list the types and level of revenue sources for each program they had in State fiscal years 1995, 1996, and 1997. We also collected information about the types of providers (private provider, local or county-owned, or State-owned) that received supplemental payments in each of the 3 years.

The survey was mailed to all States and the District of Columbia in March 1998. Forty States responded.⁵ Together, responding States accounted 89 percent of total (Federal and State) national 1997 DSH expenditures. Eleven States did not respond: Connecticut, Hawaii, Maine, Michigan, Nebraska, Nevada, New Hampshire, New Mexico, Pennsylvania, Vermont, and West Virginia. The non-responding States represent a cross-section in terms of their DSH spending relative to overall Medicaid spending—that is, some States had high levels of DSH spending while others had low levels. In 1997 DSH spending in Connecticut, Maine, Nevada, and New Hampshire accounted for more than 15 percent of total Medicaid spending in each State. By contrast, DSH spending in the remaining States ranged from less than 1 percent of program expenditures to nearly 10 percent. While we do not claim the 40 responding States are necessarily fully representative of all States, they look broadly typical.

As part of the study, we also conducted brief case studies on seven States in early 1999 to find out more details about their DSH programs and to confirm survey reportings. The study States were Alabama, California, Colorado, Louisiana, Massachusetts, Kansas, and Texas. We picked these States for several reasons. Perhaps most prominent was selecting States with large DSH programs: All of the seven States, with the exception of Kansas, had large DSH programs in which DSH expenditures accounted for at least 10 percent of the State's total Medicaid spending in 1997. We also sought to include States where the BBA DSH provisions had, or will have, varying impacts, both fiscally and programmatically. Finally, we wanted to include some States that had Medicaid provider supplemental payment programs. The case studies were designed to support the survey and help interpret the survey results.

As part of the case studies we spoke to State Medicaid officials, via telephone, to find out how their State was going to handle the BBA cutbacks in Federal DSH spending: One possible scenario would be to replace Federal dollars with State dollars. If a State had previously retained some of the Federal DSH funds, were they now going to give more to hospitals? Another issue we investigated was Medicaid supplemental payment programs. For States with such programs, we asked about the development of the program.

Results

We present our results primarily in aggregate terms. This, however, understates the complex nature of States' DSH programs. Many have more than one program which vary in both hospitals' qualifying criteria and provider payout formulae. For example, Massachusetts reported operating 11 DSH programs in 1997, while Washington reported having 9. Further, 13 States reported having a provider supplemental payment program distinct from DSH.⁶ California, Illinois, and Massachusetts reported having more than one supplemental program. Like DSH, supplemental payment programs typically varied in terms of financing and provider payout.

Medicaid DSH Revenue Sources

We classified DSH revenue sources into five groups—provider taxes, IGTs and CPEs from local or county entities, State transfers and CPEs, State general funds, and Federal Medicaid matching funds. For State funds there are fine distinctions among the various sources (State general funds or State transfers and CPEs) as they all come from the same original source—State appropriations. While here we present them as separate sources, they could be viewed as a single source, and small differences in accounting practices could alter the specific source.

As shown in Table 1, most States used some State funds to finance their programs; 28 States relied at least in part on State general funds. Only six States (California, Colorado, Indiana, Kentucky, Mississippi, and Rhode Island) relied solely on non-State (local, county, or private) funds. California is particularly noteworthy in that it relies completely on IGTs from local sources—counties, special districts and university hospitals—that operate DSH hospitals to fund the State share of DSH payments.

Table 1. Sources of Revenue for Medicaid DSH Program, by 40 Survey States: State Fiscal Year 1997.

State	Total Revenue	Revenue Sources
		Local or Private Funds		State Funds		Federal Funds
		Provider Taxes	Intergovernmental Transfers/CPEs	State Transfers/CPEs	State General Fund1	FMAP	Medicaid Matching Funds
		Thousands
Total	$15,425,642	$1,384,885	$2,934,568	$1,934,953	$1,146,696		$8,024,540
Alabama	417,500	5,146	68,043	18,208	35,773	69.54	290,330
Alaska	12,500	—	—	—	6,250	50.00	6,250
Arizona	142,364	—	—	—	49,073	65.53	93,291
Arkansas	2,569	—	—	—	686	73.29	1,882
California	2,138,571	—	1,179,774	—	—	50.23	958,797
Colorado	153,166	80,444	—	—	—	52.32	72,722
Delaware	8,870	—	—	—	4,435	50.00	4,435
District of Columbia	89,580	—	—	—	44,790	50.00	44,790
Florida	365,793	—	71,794	80,324	9,599	55.79	204,076
Georgia	430,139	—	170,609	8,172	—	61.52	251,358
Idaho	2,552	—	—	—	817	67.97	1,735
Illinois	477,900	—	—	184,250	54,700	50.00	238,950
Indiana	109,043	—	45,000	—	—	61.58	64,043
Iowa	14,886	—	—	—	5,517	62.94	9,370
Kansas	61,900	—	—	—	25,459	58.87	36,441
Kentucky	238,856	94,022	—	—	—	70.09	144,834
Louisiana	662,995	—	—	—	189,882	71.36	473,113
Maryland	160,470	—	—	61,263	18,972	50.00	80,235
Massachusetts	805,500	315,000	36,000	165,000	—	50.00	289,500
Minnesota	60,935	—	—	—	28,274	53.60	32,661
Mississippi	287,250	—	137,722	—	—	77.22	149,528
Missouri	832,817	325,725	31,869	74,285	—	60.04	400,939
Montana	250	—	—	—	77	69.01	173
New Jersey	966,600	—	43,500	216,900	222,900	50.00	483,300
New York	2,974,055	300,808	578,280	281,237	326,703	50.00	1,487,027
North Carolina	361,662	—	125,996	13,990	—	63.89	221,676
North Dakota	1,297	—	—	—	419	67.73	878
Ohio	682,393	226,278	—	44,171	7,422	59.28	404,523
Oklahoma	25,748	—	—	7,397	325	70.01	18,026
Oregon	31,968	—	—	—	12,621	60.52	19,347
Rhode Island	70,106	37,462	—	—	—	53.90	32,644
South Carolina	445,521	—	114,981	14,280	2,480	70.43	313,780
South Dakota	1,570	—	—	—	551	64.89	1,019
Tennessee2	-	—	—	—	—	64.58	—
Texas	1,907,533	—	331,000	630,000	—	62.56	946,533
Utah	5,378	—	—	477	1,011	72.33	3,890
Virginia	132,146	—	—	—	64,157	51.45	67,989
Washington	331,000	—	—	135,000	28,779	50.52	167,221
Wisconsin	12,154	—	—	—	4,983	59.00	7,171
Wyoming	105	—	—	—	42	59.88	63

¹States were inconsistent in their reporting of State general funds. Therefore, these revenues were imputed based on the assumption that, given their DSH payments and all other sources of revenue, States used only the amount of general fund monies needed to meet their State match.

²Tennessee was a reporting State. However, in their response they stated they no longer maintained a DSH program as DSH payments are included in their managed care capitation rates used in their section 1115 TennCare waiver program.

NOTES: DSH is disproportionate share hospitals. CPE is certified public expenditures. FMAP is Federal medical assistance percentage.

SOURCE: The Urban Institute Survey of State Medicaid DSH and other Payment Programs, 1999.

Many States, including New York, Alabama, Florida, and Georgia, used a combination of revenue sources to support the program. Despite the 1991 Federal DSH legislation which greatly limited States' use of provider taxes, eight States still use tax revenues to at least partially support their program.

In 1997 the 40 reporting States collected a total of $15.4 billion in revenues to support their DSH programs. As expected Federal matching funds were the largest funding source, accounting for about one-half (52 percent) of the revenues or $8 billion. Among the States' share, IGTs and CPEs from county and local entities were by far the largest funding source, contributing about $2.9 billion (19 percent). Nearly threequarters of this is accounted for by large IGT/CPE programs in California, New York, and Texas. Fund transfers or CPEs from State entities were the second largest funding source for the States' share at $1.9 billion (13 percent), followed by provider taxes at $1.14 billion (9 percent). State general funds accounted for $1.1 billion (7 percent).

DSH Payments

Table 2 shows the distribution of total (Federal and State) DSH payments by type of hospital (acute or mental) and by ownership (private, local or county, and State). As shown in Table 1, the 40 survey States made a total of $14.2 billion in DSH payments in 1997. This is $1.2 billion less than what States received in DSH revenue. Table 3 shows that this occurred because States did not always pay out in DSH payments all that they took in.

Table 2. Disproportionate Share Hospital (DSH) Payments, by Type of Hospital and Ownership Status for 40 Survey States: State Fiscal Year 1997.

State	DSH Payments
	Total DSH Payments	Acute Hospitals			Mental Hospitals
		Private	Local/County Public	State	Private	Local/County Public	State
		Thousands
Total	$14,218,981	$3,733,479	$5,161,561	$2,295,754	$97,980	$91,569	$2,838,638
Alabama	417,500	175,950	212,450	—	—	—	29,100
Alaska	12,500	—	—	—	500	—	12,000
Arizona	142,364	17,848	97,410	—	—	—	27,107
Arkansas	2,569	1,883	2	139	543	—	—
California	1,908,813	365,352	1,393,003	146,153	4,306	—	—
Colorado	138,995	15,759	3,408	119,5171	57	—	254
Delaware	8,870	—	—	—	—	—	8,870
District of Columbia	89,580	41,986	—	23,797	—	—	23,797
Florida	365,793	58,570	122,647	33	—	—	184,543
Georgia	408,580	47,346	300,802	60,432	—	—	—
Idaho	2,552	1,492	913	147	—	—	—
Illinois	477,900	68,100	1,400	82,300	1,800	—	324,300
Indiana	104,000	—	—	104,000	—	—	—
Iowa	14,886	727	338	13,822	—	—	—
Kansas	61,900	3,000	1,900	1,600	—	—	55,400
Kentucky	206,640	87,689	—	65,717	2,438	—	50,795
Louisiana	662,995	5,364	5,198	562,218	452	—	89,762
Maryland	160,470	37,768	—	175	—	—	122,527
Massachusetts	579,000	183,000	231,000	48,000	—	—	117,000
Minnesota	60,935	49,734	—	—	—	—	11,201
Mississippi	193,639	2,359	121,960	69,320	—	—	—
Missouri2	667,786	193,292	267,227	34,366	9,813	—	163,088
Montana	250	—	250	—	—	—	—
New Jersey	966,600	407,000	29,600	180,300	—	86,900	262,800
New York	2,974,055	1,261,469	1,015,575	101,402	70,574	—	525,035
North Carolina	346,965	—	189,295	17,610	—	—	140,060
North Dakota	1,297	205	—	—	—	—	1,092
Ohio	682,393	258,263	258,263	57,392	—	—	108,475
Oklahoma	25,748	434	—	21,365	649	—	3,300
Oregon	31,968	546	—	4,548	—	—	26,874
Rhode Island	60,564	46,661	—	13,903	—	—	—
South Carolina	445,521	102,545	293,783	10,596	903	—	37,695
South Dakota	1,570	289	30	—	—	—	1,250
Tennessee3	—	—	—	—	—	—	—
Texas	1,513,000	263,000	615,000	236,000	5,000	—	394,000
Utah	5,378	1,716	—	2,494	—	—	1,168
Virginia	132,146	15,450	—	110,406	780	—	5,511
Washington	331,000	13,000	—	208,000	—	—	110,000
Wisconsin	12,154	5,684	—	—	166	4,669	1,635
Wyoming	105	—	105	—	—	—	—

¹Colorado's DSH payments to State hospitals include payments to quasi-governmental hospitals—most prominently Denver Health and Hospital Authority.

²Missouri reported $470,332 in DSH payments to “non-State” hospitals, without breaking payments into type or ownership categories. In our analysis, we distributed these non-State DSH payments to match the National Distribution of DSH payments.

³Tennessee was a reporting State. However, in their response they stated they no longer maintained a DSH program as DSH payments are included in their managed care capitation rates used in their section 1115 TennCare waiver program.

SOURCE: The Urban Institute, 1999.

Table 3. Distribution of Medicaid Inpatient Days and Medicaid Disproportionate Share Hospital (DSH) Payments, by Type of Hospital for 40 Survey States: 1996 and 1997.

Type of Hospital	Total	Acute Care Percent	Mental Health
1996 Inpatient Days
Total	100.0	93.6	6.4
State-Owned	7.0	4.9	2.1
Local/County Public	24.5	23.7	0.8
Private/Non-Profit	68.5	65.0	3.5
1997 DSH Payments
Total	100.0	78.8	21.2
State-Owned	36.1	16.1	20.0
Local/County Public	38.4	37.7	0.6
Private/Non-Profit	25.5	24.9	0.6

NOTE: The American Hospital Association data exclude Federal hospitals and hospitals that are not classified as acute care or psychiatric hospitals.

SOURCES: Urban Institute DSH Survey and analyses of American Hospital Association Annual Survey of Hospitals, 1996.

Among the $14.2 billion in DSH payments, acute care hospitals received the bulk (79 percent) of the funds or about $11.2 billion. Of payments going to acute care hospitals, local and county public hospitals facilities received the most—almost $5.2 billion (36 percent of the total). Private acute care hospitals (non-profit and for-profit) followed with $3.7 billion (26 percent) and State acute care hospitals received $2.3 billion (16 percent).

Most States made payments to all three types of acute care hospitals. Excluding Tennessee, which reported making no DSH payments, all but a few States (Alaska, Delaware, Indiana, Montana, North Carolina, and Wyoming) made DSH payments to private acute care hospitals. Further, all but 15 States issued DSH payments to local and county hospitals. Among these 15 States, in 4 there were no local or county hospitals. As to State acute hospitals, all but 10 States made payments to State acute hospitals. Among these 10, in 6 there were no State acute care hospitals.

Overall, mental hospitals received 21 percent of DSH payments in 1997. All but eight States—Georgia, Idaho, Indiana, Iowa, Mississippi, Montana, Rhode Island, and Wyoming—reported payments to IMDs.⁷ Among the States that did issue IMD DSH payments, State-owned facilities received almost all the payments. In six States—Florida, Illinois, Kansas, Maryland, Oregon, and South Dakota—IMDs received more than one-half the overall DSH expenditures. As previously mentioned, using DSH funds to support mental hospitals has been particularly troubling in the eyes of Federal policymakers; the BBA ratchets down the extent to which DSH can be paid to IMDs over time.

In terms of hospital ownership, local and county hospitals (acute and mental) received the most DSH dollars, $5.2 billion or roughly 37 percent of total payments. State-owned hospitals followed closely with $5.1 billion, about 36 percent of overall payments. In some States (including Louisiana, Colorado, Washington, and Virginia) State acute-care hospitals received the bulk of the DSH funds. In Louisiana and Washington, for example, these acute-care hospitals were State university hospitals. Finally, private hospitals received a total of $3.6 billion or about 25 percent of total payments.

Table 2 also shows that States generally pay DSH to more than one type of provider. Several of the larger States—Illinois, New York, and Texas—make payments to five types of providers. With the exception of Montana and Wyoming, all States made DSH payments to State facilities, acute or mental.

Medicaid DSH programs are often designed to provide higher payments to certain types of hospitals. As one measure of the relative funding provided by hospital type, Table 3 compares the distribution of Medicaid DSH payments and the distribution of Medicaid inpatient days, as reported in the American Hospital Association's annual survey of hospitals (American Hospital Association, 1996). It should be noted that we are relying on just Medicaid inpatient days. We do not account for days of care provided to the uninsured, which are a key component to hospitals' uncompensated care costs. For example, State-owned hospitals (acute care and psychiatric combined) provide 7 percent of the care to Medicaid patients (based on inpatient days), but receive 36 percent of DSH payments. By contrast, local/county public hospitals and private/non-profit hospitals provide 25 and 69 percent of all Medicaid inpatient days, but received 38 and 26 percent of DSH payments, respectively.⁸ While we cannot account for the distribution of uncompensated care, it seems clear that State hospitals receive an uneven share of DSH payments.

A comparison of Medicaid inpatient volume and DSH payments also shows that DSH programs favor IMDs. Although psychiatric hospitals provide only 7 percent of Medicaid inpatient days, they receive 21 percent of Medicaid DSH payments. As previously noted, Medicaid does not cover inpatient psychiatric care for adults.

Net Gains Through the Program

By comparing the revenues contributed and actual DSH payments, we can compute how much the hospitals and the States gain through DSH. DSH gains can be achieved by two ways—through hospitals or State residual funds. As for hospital gains, we first grouped hospitals (acute and mental) by ownership status—State and non-State (private and local/county). To calculate hospitals' net gain, we assumed that hospitals, county or local governments, and the State have been paid back in full any funds they may have contributed for the State share of its DSH program. For example, assume that a county hospital transferred $100 million in IGTs to support the DSH program. The State then made a $120-million DSH payment to this hospital. As shown in Table 4, the net gain for that hospital would be listed as $20 million ($120 million payment minus $100 million IGT). By making this assumption, we infer that whatever funds (taxes, IGTs, CPEs, or State transfers) providers or other entities submit to the State for DSH will be paid back to the contributor and do not represent new funds available to hospitals. For different hospital sectors (private, county, State and the like) this assumption is sound. However, at the individual hospital level there may be winners and losers under the DSH program, especially if provider taxes are the funding source.

Table 4. Net Disproportionate Share Hospital (DSH) Gains by Hospital Ownership¹ and DSH Residual Funds for 40 Survey States: State Fiscal Year 1997.

State	Net DSH Gains				Percent of DSH Gains to State3
	Total	Gains to Private or Local Entities by DSH Payments	Gains to State Entities Through DSH Payments2	State Gains from Residual Funds
		Thousands
Total	$8,024,540	$4,765,136	$2,052,742	$1,206,661	40.6
Alabama	290,330	315,210	-24,881	—	-8.6
Alaska	6,250	500	5,750	—	92.0
Arizona	93,291	115,257	-21,966	—	-23.5
Arkansas	1,882	2,429	-547	—	-29.0
California	958,797	582,886	146,153	229,758	39.2
Colorado	72,722	-61,221	119,771	14,171	184.2
Delaware	4,435	—	4,435	—	100.0
District of Columbia	44,790	41,986	2,804	—	6.3
Florida	204,076	109,424	94,652	—	46.4
Georgia	251,358	177,539	52,260	21,559	29.4
Idaho	1,735	2,405	-671	—	-38.7
Illinois	238,950	71,300	167,650	—	70.2
Indiana	64,043	-45,000	104,000	5,043	170.3
Iowa	9,370	1,065	8,305	—	88.6
Kansas	36,441	4,900	31,541	—	86.6
Kentucky	144,834	-3,895	116,512	32,216	102.7
Louisiana	473,113	11,014	462,099	—	97.7
Maryland	80,235	37,768	42,467	—	52.9
Massachusetts	289,500	63,000	—	226,500	78.2
Minnesota	32,661	49,734	-17,073	—	-52.3
Mississippi	149,528	-13,403	69,320	93,611	109.0
Missouri	400,939	112,738	123,169	165,031	71.9
Montana	173	250	-77	—	-44.9
New Jersey	483,300	480,000	3,300	—	0.7
New York	1,487,027	1,468,531	18,497	—	1.2
North Carolina	221,676	63,300	143,680	14,697	71.4
North Dakota	878	205	674	—	76.7
Ohio	404,523	290,249	114,274	—	28.2
Oklahoma	18,026	1,083	16,943	—	94.0
Oregon	19,347	546	18,801	—	97.2
Rhode Island	32,644	9,199	13,903	9,542	71.8
South Carolina	313,780	282,249	31,531	—	10.0
South Dakota	1,019	320	699	—	68.6
Tennessee4	—	—	—	—	—
Texas	946,533	552,000	—	394,533	41.7
Utah	3,890	1,716	2,174	—	55.9
Virginia	67,989	16,229	51,760	—	76.1
Washington	167,221	13,000	154,221	—	92.2
Wisconsin	7,171	10,519	-3,348	—	-46.7
Wyoming	63	105	-42	—	-67.0

¹For each ownership status type, DSH payments to both acute and mental hospitals are combined.

²These net gains were calculated by subtracting State transfers and State general fund revenues from total payments made to State providers.

³These are the sum of State gains through DSH payments and residual funds.

⁴Tennessee was a reporting State. However, in their response they stated they no longer maintained a DSH program as DSH payments are included in their managed care capitation rates used in their section 1115 TennCare waiver program.

SOURCE: The Urban Institute, 1999.

In the example previously cited, the State gained $40 million. Assuming the State had a 50-percent match rate, the State would receive $60 million in Federal DSH funds. Of the $60 million, $20 million went to the hospital and $40 million was retained by the State. We call the money retained by the State residual funds. The level and distribution of residual funds among States is shown in Table 4.

We calculated that, collectively, States and hospitals gained a total $8 billion through the DSH program (Table 4). Non-State hospitals (private and county/local hospitals) netted $4.8 billion or about 60 percent of the total possible gains. State hospitals gained about $2 billion. We also estimated that States kept $1.2 billion in DSH residual funds. Eleven States reported collecting more than they paid out. Some maintained fairly large residual funds. The bulk of the residual funds—$850 million out of the $1.2 billion total—is concentrated in three States—California, Massachusetts, and Texas. For example, in 1997, California kept $229 million in residual funds as a fee to administer the DSH program. Between residual DSH funds and State hospital gains, we estimated that State entities netted nearly $3.3 billion from the DSH program in 1997, 40 percent of the total possible gains.

About 40 percent of DSH gains went to the State, either through residual funds or through DSH payments. However, this statistic varied greatly across the States. Several States, such as Colorado, Delaware, Indiana, Louisiana, and Mississippi, kept all of the gains. Other States kept the bulk of the gains. These included Alaska, Illinois, Iowa, Kansas, Oregon, and Washington. The States that provided large payments to IMDs typically had very large State gains, since these payments were primarily aiding State mental hospitals. By contrast, a few States—Alabama, Idaho, and Wisconsin—had little or no gain or even paid more in State funds than they received back.

When public hospitals have net gains, it is possible that they do not get to keep their gains. That is, the net DSH gains experienced by State hospitals may be returned to the general State treasury. Based on the interviews conducted for this study and the earlier one (Ku and Coughlin, 1995), our impression is that State DSH hospitals gains typically do not represent new additional funds. Instead, States often reduce other subsidies to those hospitals that offset the DSH gains. Reducing existing subsidies may also occur with local or county public hospitals, that is, their gains may revert back to the county or city treasury. However, we believe this is somewhat less common at the county/local level and data from the National Association of Public Hospitals indicate how DSH has helped the overall operating margins of their members, which are primarily local public hospitals (Fagnani and Tolbert, 1999).

Because of the fungibility of funds and variations in accounting practices, it is difficult to determine the extent to which additional funds earned by public hospitals in DSH actually result in providing more health care versus merely reducing the need for other State or local revenue for the hospital. For example, suppose that a public hospital has a deficit of $10 million without accounting for DSH, but gets $10 million in DSH net gains. If the county or State would otherwise cover the deficit with other public monies, then the DSH gains can be viewed as simply replacing other funds and the true beneficiary is the State or local treasury. On the other hand, if the county or State forced the hospital to cover its deficit by reducing services in the next fiscal year, then the $10 million from DSH could be viewed as helping the hospital to provide more services. Based on our study, it was not possible to determine the extent to which DSH payments act as substitutes or supplements to other public funding of hospitals.

Distribution of Federal DSH Payments Among States

A key issue for congressional policymakers has been whether DSH funding is allocated equitably across States. The 1991 and 1997 laws explicitly capped DSH allocations for States and particularly sought to restrain high DSH States. In this analysis, we develop and compare two parameters that try to measure the relative need for DSH across States.

As shown in Table 5, the first parameter is the level of total Medicaid DSH payments per uninsured person and Medicaid recipient. This measures the DSH payments in each State, relative to the size of its number of uninsured and Medicaid population, the intended patient pool to be aided under the DSH program. Nationally, DSH payments were $218 per uninsured and Medicaid recipient in 1997. However, for individual States the payment varied greatly, ranging from a less than $1 in Wyoming, West Virginia, Nebraska, and Hawaii to nearly $700 in Connecticut.

Table 5. Total Disproportionate Share Hospital (DSH) Dollars per Uninsured or Medicaid Individual and Total DSH Dollar, by State: Federal Fiscal Year 1997.

State	Total Medicaid DSH1 per Medicaid or Uninsured Individual2	Total Medicaid DSH1 per Medicare DSH3
United States	$218.96	$3.56
Alabama	391.89	4.85
Alaska	80.86	4.29
Arizona	82.65	3.40
Arkansas	3.11	0.10
California	185.11	2.71
Colorado	173.57	5.99
Connecticut	689.73	12.78
Delaware	56.75	1.36
District of Columbia	225.48	1.37
Florida	85.68	1.27
Georgia	178.83	2.77
Hawaii	0.00	0.00
Idaho	18.48	0.91
Illinois	105.97	1.43
Indiana	183.25	4.19
Iowa	31.96	1.48
Kansas	125.84	3.64
Kentucky	208.71	3.87
Louisiana	423.64	5.16
Maine	566.31	11.33
Maryland	156.21	—
Massachusetts	372.33	7.33
Michigan	136.07	2.55
Minnesota	53.15	1.58
Mississippi	228.35	3.02
Missouri	587.11	11.99
Montana	0.83	0.05
Nebraska	0.00	0.00
Nevada	179.42	4.36
New Hampshire	647.87	—
New Jersey	525.60	10.10
New Mexico	14.91	0.65
New York	472.43	4.28
North Carolina	191.02	1.76
North Dakota	15.89	0.91
Ohio	303.48	6.15
Oklahoma	27.49	0.74
Oregon	30.96	1.40
Pennsylvania	252.59	2.92
Rhode Island	307.37	7.67
South Carolina	442.07	5.11
South Dakota	9.98	0.56
Tennessee	0.21	0.00
Texas	218.85	4.00
Utah	14.44	0.61
Vermont	225.30	11.92
Virginia	114.98	2.30
Washington	257.09	7.09
West Virginia	0.00	0.00
Wisconsin	14.21	0.35
Wyoming	0.00	—

¹Total Medicaid DSH dollars are from HCFA-64 reports submitted by States in Federal fiscal year 1997.

²The number of Medicaid and uninsured individuals was estimated using a 2-year merged sample of 1997 and 1998 March Current Population Surveys.

³Medicare DSH dollars are from the Prospective Payment Assessment Commission (ProPAC), 1997.

NOTE: Dash indicates that the denominator is zero.

SOURCE: The Urban Institute, 1999.

The second measure tested is the ratio of Medicaid DSH to Medicare DSH payments in each State. Unlike Medicaid DSH, Medicare DSH payments are based on standardized national formulae. This ratio also shows tremendous variety among the States, going from a low of zero to high of about 13, with the national average about 3.5.

In general, the relative rankings of States as high or low are comparable for the two measures.⁹ This uneven distribution of DSH dollars among States has been a consistent pattern in the program since the early 1990s.

Comparison of 1993 and 1997 Surveys

We compared our 1997 survey findings to a 1993 survey on State DSH programs (Ku and Coughlin, 1995). Of the 31 States that responded to both surveys; only results from these States are reported in Table 6.¹⁰ The top panel lists revenue sources in each of the two survey years. As shown, the level of revenue collected by the 31 States was about the same in both 1993 and 1997. However, the distribution of revenues changed dramatically between the two survey years. In 1993, 25 percent of DSH revenues came from provider taxes and donations. By 1997, provider taxes accounted for 10 percent of DSH revenues. This shift corroborates other findings and is primarily due to the 1991 Federal legislation that limited States' use of provider taxes and required that taxes be broad-based, and not directly linked to Medicaid or to DSH payments received.

Table 6. DSH Revenues and Expenditures, a 31-State Comparison: State Fiscal Years 1993 and 1997.

DSH Revenues and Expenditures	1993	1997	1993	1997	Average Annual Growth 1993-1997
	Millions		Percent
Revenues	13,612	12,900	100.0	100.0	-1.3
Provider Taxes and Donations	3,334	1,291	24.5	10.0	-21.1
County/Local Funds 1	1,858	2,457	13.6	19.0	7.2
State Funds 2	1,702	2,521	12.5	19.5	10.3
Federal Matching Payments	6,718	6,631	49.4	51.4	-0.3
Expenditures	13,612	12,900	100.0	100.0	-1.3
Payback to Private and County/Local	5,192	3,748	38.1	29.1	-7.8
Private and County/Local Gain	2,394	3,904	17.6	30.3	13.0
Payback to State	1,702	2,521	12.5	19.5	10.3
State Hospital Gain	2,439	1,682	17.9	13.0	-8.9
Residual Funds for State Use	1,885	1,045	13.8	8.1	-13.7

¹County/local funds include both intergovernmental transfers and certified public expenditures from county or local hospitals (acute and mental) or entities.

²State funds include both certified public expenditures or State transfers from State hospitals (acute and mental) or entities.

SOURCE: The Urban Institute Survey of State Medicaid DSH and Other Payment Programs, 1999 and Ku and Coughlin, 1995.

As States decreased use of provider taxes, their reliance on IGT/CPEs from county and local entities increased. Between 1993 and 1997, the percent of DSH revenues obtained from county and local funds went from 14 to 19 percent. The jump in State funds—general funds, CPEs, or State transfers—as a revenue source was greater, going from 13 percent in 1993 to 20 percent in 1997. Federal matching payments remained about the same in both years.

Table 6 also gives an account of DSH expenditures in the 2 survey years. The 31 States spent about $13 billion in each of the 2 years, accounting for about 70 percent of national DSH expenditures in 1993 and 74 percent in 1997. While the level of payments remained the same, how States distributed DSH funds changed over the 4 years. Most notable is the increase in gains to private and county/local hospitals: In 1993 these hospitals gained about $2.4 billion. By 1997 their gains were estimated at nearly $3.9 billion. Another important difference between the 2 survey years is the drop in State hospital DSH gains. Between 1993 and 1997, State hospital gains went from $2.4 billion to $1.7 billion. Likewise, the amount of DSH residual funds kept by the State declined, going from about $1.9 billion in 1993 to $1 billion in 1997. Thus, total State DSH gains went from $4.3 billion in 1993 down to $2.7 billion in 1997. In short, this comparison suggests that an increasing share of the available DSH funds to be gained is going to private and county/local hospitals while States are keeping less.

This distributional shift was caused by changes in both the revenue and expenditure streams of States' DSH programs. First, provider tax revenue declined by 1997. As provider tax revenue fell, States relied more on State or county transfers as revenue sources. Second, because of the 1993 legislation that capped DSH payment levels to 100 percent of uncompensated costs and required a 1-percent Medicaid volume, States were forced to limit the level of DSH expenditures paid to State hospitals and IMDs. Together, the increase in State revenue contributions and the leveling off of payments to State facilities caused State effective gains through the DSH program to fall, while those of local and county hospitals rose.

Early Impacts of BBA DSH Provisions

DSH financing and provider payout will change yet again because of the 1997 BBA DSH provisions. As part of our seven-State case studies we asked State officials how BBA has affected them to date and how they intended to handle future cutbacks. As to the first issue, in general, officials in the seven study States said that so far, the BBA had not posed a significant hardship. In large measure, this is because some of the States' largest cutbacks and the phase-in of the IMD limitations do not occur until 2000. As such, at the time of our case studies the financial reality of the DSH cutback, had not yet been fully felt. Another reason cited was that States are currently in strong fiscal shape, so the DSH cutbacks are not so burdensome. Texas officials further noted that the State just received a multi-billion dollar settlement with the tobacco industry. As part of that settlement, hospitals will receive more than $2 billion which has helped offset some of the DSH cutbacks. (All States have had substantial gains from the tobacco settlements.) A couple of States felt that the implementation of the Children's Health Insurance Plan has also helped compensate for some of the DSH losses in that more children are now insured thus easing hospitals' uncompensated care levels.

California and Massachusetts acknowledged that, in part, some of the potential DSH pressure was relieved because they had obtained special Federal exemptions. Massachusetts received a section 1115 waiver, which, among other things, allowed the State to obtain more Medicaid funds for safety net hospitals. Federal legislation in 1999 extended California's ability to pay its hospitals up to 175 percent of uncompensated care rather than 100 percent. This exemption was sought so that county hospitals, who supply the State's DSH share through IGTs, could receive more DSH funds. In addition, in State fiscal year 1998, the State reduced its gain from the DSH program (from about $230 million to $114 million) so that more DSH funds could be paid to providers.

Colorado responded to the BBA by increasing its current DSH spending. For the past several years, the State has not fully spent its nearly $302 million DSH allotment which will be cut in half after 2000. However, in 1998, the State decided to fully spend its allotment while they could. They are viewing this as a rainy day fund for hospitals. By making DSH payments to State hospitals, Colorado plans to fully spend its allotment in 1999 and 2000.

Although our seven study States did not seem too concerned about the DSH cutbacks at present, many saw problems on the horizon and are trying to plan ahead. In 1999, the Alabama Medicaid Agency, for example, requested additional money from the legislature to partially offset some of the DSH losses. They have also increased spending on their supplemental payment programs for public hospitals. State officials, however, were not hopeful about securing additional State funds and felt that some cuts were going to be made. Louisiana is considering increasing Medicaid reimbursement rates to help offset DSH losses, but was unsure at present, how the increase would be funded. Similarly, Massachusetts officials consider expanding their supplemental payment programs as the DSH cutbacks are phased in. California officials stated that they did not see that the State would come up with additional funds to backfill the DSH losses. They did, however, see that the State may consider reducing its gain on the DSH program again as a way to offset hospitals' future DSH cutbacks.

Medicaid Supplemental Payment Programs

A 1998 study of 13 States reported that several States had recently established Medicaid supplemental payment programs besides their DSH program (Coughlin and Liska, 1998). As part of the survey we asked States whether they had any supplemental payment programs. Under such programs additional payments are made to selected providers (hospitals, as well as, other provider types) over and above regular Medicaid service reimbursement. These supplements are not counted against a State's DSH cap or a hospital's DSH cap; instead they are reported as a regular Medicaid expenditure. Thus, supplemental payment programs allow States to target additional Medicaid funds to certain providers.¹¹

Based on State's description of their programs supplemental payment programs function much like DSH programs. For example, in response to the 1993 hospital-specific DSH caps, Alabama implemented its Public Hospital Enhancement program in 1994. The State's share of the program is funded by IGTs from public hospitals that are transferred to the State Medicaid agency. The State then pays the public hospitals two types of enhancements. The first is for inpatient care in which an enhancement is made to public hospitals for each covered Medicaid day. The enhancement is the difference between what Medicaid pays and the Medicare upper payment limit. The program also enhances Medicaid outpatient payments for public hospitals in the same way. In 1998, it is estimated that Alabama public hospitals received a total of $125 million in enhancements. According to State officials, Alabama established the enhancement program as a way to get Medicaid dollars to public hospitals as the hospital-specific caps included in 1993 law were phased in.

Using a somewhat different approach to secure additional Federal Medicaid funding for public hospitals, in 1997 Massachusetts started making supplemental payments to two public hospitals—Cambridge Hospital and Boston Medical Center. More specifically, as part of the State's section 1115 waiver program, enhanced capitation payments are paid to the health plans operated by the these hospitals. In 1997, an estimated payment of $193 million was made to the two hospitals. According to State officials, these supplemental payments are to provide financial support to the hospitals as they move to the more competitive managed care environment. In addition to this supplemental payment program, Massachusetts also makes supplemental payments to home health agencies and freestanding community health centers.

Among the 40 States responding to our survey, 13 had at least one supplemental payment program in 1997 (Table 7). Several States—including California, Massachusetts, and Illinois—reported having more than one. While supplemental payment programs are made largely to hospitals, States did report making such payments to other providers such as home health agencies and nursing homes. As shown in Table 7, most ($900 million) of the States' share for supplemental payment programs comes from non-State sources, largely IGTs and CPEs from county and local entities. State revenue accounted less than $1 million. The Federal Government paid $932 million in Federal matching payments to support supplemental programs in 1997.

Table 7. Revenues for Supplemental Provider Payment Programs for 40 Survey States, State Fiscal Year 1997.

State	Non-State Revenue	State Revenue	Federal Revenue
	Thousands
Total	$901,815	$77,869	$932,419
Alabama	21,340	—	42,211
California	332,201	28,792	364,329
Florida	—	975	544
Illinois	369,700	—	297,950
Indiana	35,000	—	9,237
Massachusetts	16,000	—	8,000
Minnesota	—	4,333	5,005
New York	127,574	—	127,574
North Carolina	—	43,278	76,572
Oregon	—	—	—
South Carolina	—	296	704
Wisconsin	—	—	—
Wyoming	—	196	293

¹Although Wisconsin responded to our survey, it did not respond to questions on supplemental payment programs. However, it has been reported elsewhere that Wisconsin had at least one supplemental payment program in 1997 (Coughlin and Liska, 1998).

Source: The Urban Institute, 1999.

Supplemental payment program spending has increased significantly in recent years, going from $0.6 billion in 1995 to about $2.8 billion in 1998 (Table 8). The bulk of the funding for supplemental payment programs came from non-State entities, largely through IGTs or CPEs from county or local providers. In 1998, for example, non-State entities provided $1.3 billion to support supplemental payment programs.

Table 8. Supplemental Provider Payment Programs for 40 Survey States State Fiscal Years 1995-1998.

Revenues and Expenditures	1995	1996	1997	1998
	Thousands
Revenues	$643,466	$1,918,035	$1,912,102	$2,775,475
Non-State Funds1	282,169	995,909	901,815	1,323,233
State Funds2	47,466	6,358	77,869	64,422
Federal Matching Payments	313,830	915,768	932,419	1,387,821
Expenditures	643,466	1,918,035	1,912,102	2,775,475
Payback to Non-State Providers3	282,169	995,909	901,815	1,323,233
Gain to Non-State Providers	303,780	809,558	897,417	1,338,467
Payback to State Providers4	47,466	6,358	77,869	64,422
Gain/Loss to State Providers	-47,466	-6,358	-77,380	-63,921
Residual Funds for State Use	57,516	112,567	112,382	113,274

¹Non-state revenues include provider taxes and intergovernmental transfers or certified public expenditures from local or county entities.

²State revenues include State general funds, State transfers or certified public expenditures from State entities.

³Payments to non-State providers include payments to privately-owned providers as well as local- or county-owned providers.

⁴State providers include payments to State-owned providers.

SOURCE: The Urban Institute Survey of State Medicaid DSH and Other Payment Programs, 1999.

The one important distinction between supplemental payment programs and the DSH program is that most of the funds paid out through supplemental programs are paid to non-State providers; only a small share of the funds are retained by the States. In 1998, for example, States reported making nearly $2.8 billion in supplemental payments. Of these funds, $2.7 billion are paid out to private and county/local providers, with about one-half ($1.3 billion) in payback and one-half ($1.3 billion) in provider gains. States' gains (State hospital gain plus residual funds) on supplemental payment programs totaled less than $.05 billion. The limited State gain under supplemental payment programs has been consistent during the period from 1995 to 1998. So while supplemental payment programs are DSH-like in many ways, our survey results suggest that, unlike DSH, virtually all of the Federal Medicaid match—and any net gains—on these programs is being paid to providers. In other words, supplemental payment programs represent a new revenue source for selected providers.

Discussion and Conclusion

In 1991, 1993, and 1997, there were major amendments to Federal legislation governing the Medicaid DSH program. Given the importance of DSH and the controversy surrounding the program, it is likely the program will continue to evolve and be subject to future amendments. Our analyses indicate that the 1991 and 1993 provisions were partially successful in reducing some of the problems associated with the DSH program. An increasing share of DSH gains is now been paid to hospitals, for example, among the 31-State comparison sample, private and county/local hospital gains went from $2.4 billion in 1993 to nearly $3.9 billion in 1997. By 1997, these hospitals netted about 60 percent of the total available DSH gains, up from 36 percent in 1993.

The increase in DSH gains by hospitals as DSH payments represent an important source of revenue for safety net hospitals. One recent study of urban public hospitals found that Medicaid DSH payments financed about 25 percent of the hospital's uncompensated care costs (Fagnani and Tolbert, 1999). The same study, as previously mentioned, reported that safety net hospitals over the past decade have become increasingly reliant on Medicaid DSH payments to help fund uncompensated care. In part this reflects the growth in the Medicaid DSH program in the 1990s. It also likely reflects the trend that uncompensated care costs have become more concentrated among fewer hospitals in recent years (Cunningham and Hu, 1997).

Despite hospitals receiving more of the DSH funds, States still retained about 40 percent ($2.7 billion) of the DSH gains in 1997. While some of these monies go to State acute-care hospitals, such as university hospitals, much of it goes to State psychiatric hospitals or kept by States for other purposes. In some instances, the DSH gains are used to support the State share of Medicaid or other health services, whereas in others they may directly support completely unrelated efforts.

Our analyses also showed that States are increasingly relying on IGT/CPEs (from local or county entities or State transfers or CPEs) to finance their share of the DSH program. While CPE/IGTs are reported as DSH revenues, we have assumed that they do not represent new funds to hospitals that can be used to provide care to the uninsured or purchase new equipment. Instead, we have viewed them as a way for States to secure additional Federal DSH matching dollars. Based on survey data of the 40 responding States, we estimated that in 1997 only about 56 percent (or $8.0 billion) of the reported $14.2 billion in DSH payments represent new additional funds that can be netted by hospitals. To calculate this, we assumed that State revenue sources, for the most part, do not represent new resources available to the health care safety net. To the extent that this assumption is incorrect and the State's share do represent new monies available to hospitals, the 56-percent figure understates possible gains through the DSH program.

Another area of the program that did not change is the distribution of Federal DSH funds among States. In 1997, total DSH payments per Medicaid and uninsured person ranged from a less than $1 to a high of nearly $700. It is too early to assess the effect of 1997 legislation, but the measure should continue to reduce program spending and reduce certain components, such as DSH payments to IMDs. However, some States have already found new ways to secure additional Medicaid funds in a DSH-like fashion via supplemental payment programs. Our data suggest that these programs are getting new funds to providers, with hospitals being the largest beneficiaries of the programs.¹² While supplemental payments may help offset some of the BBA DSH cutbacks, these new programs may raise new legislative or regulatory issues for Federal policymakers.

Thus, while the recent Federal laws have brought about changes in the DSH program, and have helped move the program closer to its original intent, several problems persist. An important question is: What future changes should be considered for the DSH program? One possibility is to look at the various proposals that have been made to reform Medicare's DSH program, such as establishing clearer national standards for DSH payment eligibility and for the allocation of DSH payments (MedPAC, 1999). Such reforms could enable hospitals with the same share of low-income patients to have comparable levels of DSH payments. National DSH standards could also reduce some of the interstate differences in DSH spending and make it more difficult to link DSH payments to revenue programs, such as intergovernmental transfers. They also may help direct the payments to those hospitals under the most financial stress, particularly important given the DSH cutbacks included in the BBA. However, the obstacles to passing national DSH standards are great because the wide variation in States' use of the DSH program. That is, there would be some winners and losers among States if national standards were adopted.

Another current proposal for the Medicare DSH program that might apply to the Medicaid program is to require States to include the costs of providing both inpatient and outpatient services to low-income patients to determine hospital eligibility for DSH payments, as well as, in their DSH distribution formulas. While States have the option to do this, many do not include outpatient costs. (Fagnani and Tolbert, 1999). This reform is appealing for several reasons. There is some evidence that safety net hospitals provide an even greater share of uncompensated care on an outpatient basis than on an inpatient basis. For example, 1997 data from members of National Association of Public Hospitals indicate that while 26 percent of their inpatient volume was for uninsured patients, 41 percent of outpatient visits were rendered to uninsured persons (Tolbert, 1999). By tying DSH payments to outpatient services as well as inpatient services would help hospitals that provide substantial amounts of uncompensated outpatient care. An even more important reason to include outpatient care in determining DSH payments is to encourage hospitals to shift care to less costly outpatient settings, where possible. If hospitals earn DSH payments only for inpatient care, they will have inappropriate incentives for inpatient services. This is particularly relevant in light of the expansion of Medicaid managed care in most States.

With the number of uninsured Americans at more than 40 million and growing, it will become increasingly important to develop a more rational DSH program, one that better directs Medicaid DSH funds to the most financially vulnerable hospitals. Such a change will help preserve hospital access for Medicaid beneficiaries. It may also help protect the Medicaid DSH program from future Federal cutbacks.