Which of the following would be considered to be uncompensated care by a healthcare organization?

Probably the most commonly expressed concern about the emergence and growth of investor-owned health care organizations is the belief that their pursuit of profitability goals will limit or preclude them from serving patients who are unable to pay or from offering needed services that cannot be provided at a profit. Implicit in this concern is the value premise—which some dispute and which is not typical of business—that health care institutions (whether for-profit or not-for-profit) have certain social responsibilities to meet individual and community needs, even needs that cannot be met profitably. Implicit also is the assumption that for-profit and not-for-profit institutions will behave differently in this regard.

Behind these concerns lies one of the most serious problems in our patchwork system of financing health care: the 35 million people who lack adequate private health insurance coverage and who are not eligible for public programs, particularly Medicare and Medicaid.1 The care of these and many underinsured people is now largely dependent on the ability and willingness of health care institutions and individual physicians to provide uncompensated care.

Although this chapter is primarily concerned with questions of access in economically unrewarding circumstances, it should be recognized at the outset that the overall impact of for-profit providers on access to care includes their effect on access to care for people who can pay. Such access has undoubtedly been enhanced by for-profit providers purchasing facilities that are in financial trouble and that might otherwise have closed, renovating and modernizing facilities and attracting new physicians as a result, investing in new services and facilities where there is demand, and responding to patients' desires for convenient hours and locations. For-profit providers also pay taxes, which are used for such purposes as policymakers may determine. Although some tax money finances care for the indigent, plainly most federal tax revenues are used for other purposes.

This chapter deals with four negative allegations regarding the behavior of for-profit providers: that they will not serve those who cannot pay; that their ability to attract paying patients will make it increasingly difficult for other type institutions to care for those who cannot pay; that they will offer only profitable services, regardless of the community's need for other services; and that they will be more likely to close hospitals that fail to achieve economic goals. The questions of whether the traditional expectation that health care institutions contribute to the care of those unable to pay is the proper mechanism for solving access problems, and whether for-profit and not-for-profit providers should be expected to contribute equally are discussed later in this chapter.

The Problem of Care for Those Unable to Pay

A basic paradox or ambivalence in American health policy is that despite the absence of a public commitment to the availability of financing to assure access to health care for all who need it, a widespread expectation exists that hospitals will provide needed care for the millions of Americans who are unable to pay part or all of their medical bills. The expectation that hospitals will do what needs to be done and will find a way to pay for it is no doubt a residue of the history of hospitals as public or charitable institutions. Hospitals provide a substantial, if inadequate, amount of uncompensated care. Although there are wide variations among hospitals, uncompensated care (the sum of charity care and bad debts) amounted to 5.4 percent of the gross revenues of the average U.S. hospital in 1983, a total of $7.8 billion (American Hospital Association, 1985). Nevertheless, in a 1982 national survey, 15 percent of uninsured families reported that they had not obtained needed medical care during the previous year, and 4 percent reported that they had been refused care for financial reasons (Robert Wood Johnson Foundation, 1983).

To what extent, if any, does the for-profit presence exacerbate these problems? A full answer to this question would require data on the number of persons who needed services but lacked the means to pay (1) who didn't believe they could obtain services so did not try, (2) who sought care but were turned away, (3) who were given emergency care and transferred to another institution; or (4) who were cared for with the provider (hospital, nursing home, HMO, ambulatory care center, etc.) absorbing the costs in full or in part. Unfortunately, such data are largely lacking. Existing evidence is limited mostly to hospitals. Even then, it is limited to (1) survey data on the number of uninsured people treated and (2) the dollar value of uncompensated care provided. Furthermore, the measure ''uncompensated care" (charity care and bad debt as a percentage of revenues) includes the uncollected charges from people who were admitted as paying patients but whose insurance did not cover full costs, and from people who were known at the time of admission to be unable to pay. In short, available data are fragmentary and are, at best, suggestive of the true extent to which people who are unable to pay are served.

The legal responsibilities of institutions toward people who are unable to pay for their care vary, but are generally limited. Case law, accreditation requirements, and some state statutes require treatment of patients in immediately life-threatening circumstances, but this is very different from extending the full resources of a hospital to all who need care. Many public hospitals are legally obligated to care for all residents in their area. Some voluntary hospitals are supposed to provide a set amount of free care annually because of earlier receipt of Hill-Burton construction funds, but the amount of care required is limited and applies to a decreasing number of hospitals. Obligations to provide free care as a condition of the tax-exempt status of not-for-profit institutions have largely disappeared insofar as federal income taxes are concerned,2 although a 1985 Utah Supreme Court decision tied hospital exemptions from local property taxes to several criteria, including provision of charity care. Other not-for-profit hospitals that are not statutorily responsible for caring for the needy of the community nevertheless see maintaining an "open door" as a fundamental value.

The Financing of Uncompensated Care

Institutions that provide uncompensated care and unprofitable services must recover the costs elsewhere. The choices are few.

The costs can be subsidized out of revenues from paving patients, a practice that has been facilitated by third-party payers who pay hospital charges set high enough to allow such subsidization. (Bad debts for non-Medicare patients were never an allowable cost under Medicare's cost-based reimbursement rules.) However, private third-party payers (including self-insuring companies) are becoming much less willing to pay for costs incurred in the care of patients that they do not insure. Changes in the reimbursement environment—most notably in the growth of negotiated rates and price competition—are making cross-subsidization increasingly problematic. As these changes continue to happen, institutions that provide uncompensated care will increasingly have to finance such care in one of two ways—by cannibalizing themselves (using money that should go for future capital needs, deferring maintenance, cutting costs excessively, having operating deficits) or by raising money from nonpatient care sources.

Obtaining nonpatient revenues has long been essential to the sound operation of hospitals. Indeed, until recent years, the average U.S. hospital would have had a negative margin (i.e., more expenses than revenues) if it had not obtained revenues from sources other than patient care, as Table 5.1 shows. Sources of nonpatient revenues include governmental appropriations; charitable contributions; interest; and income from gift shops, parking facilities, and other subsidiary organizations. As Table 5.2 shows, the magnitude of nonpatient revenues varies across hospital types, comprising 15 percent of the total revenues of public hospitals in 1983, 5 percent of the total revenues of not-for-profit hospitals, and less than 2 percent of the total revenues of for-profit hospitals. Governmental grants and appropriations are available mainly to public hospitals. Charitable contributions are available to not-for-profit and public institutions, but now are less than 1 percent of hospital revenues.

TABLE 5.1

Revenue Margins for U.S. Community Hospitals, 1963-1984.

TABLE 5.2

Sources of Revenue, Community Hospitals by Ownership, 1983 (millions of dollars).

Nevertheless, because of past public policies—particularly the public sector's unwillingness to use tax revenues to finance care for those who are unable to pay—access to medical care for millions of people depends on the policies, practices, and resources of health care providers. It is in this context that concern about the behavior of for-profit organizations has developed.

The Providers of Uncompensated Care

In assessing the relationship between type of ownership and service to patients who are unable to pay, the committee examined national data and data from five states in which for-profit hospitals have a relatively large market share. All data sources show that public hospitals provide a disproportionately large amount of uncompensated care relative to their gross patient revenues. But since the data are generally not adjusted to reflect the governmental appropriations that such institutions receive, data from public hospitals are not strictly comparable to private institutions. A similar point pertains to comparisons of for-profit and not-for-profit hospitals: to the extent that not-for-profit hospitals receive more nonpatient care revenues (via philanthropy, governmental grants, or other sources) than do for-profit hospitals, provision of uncompensated care requires less subsidization from patient care revenues.

The most recent national data on who provides uncompensated care come from two sources: a 1981 survey conducted by the Office for Civil Rights (OCR) in the Department of Health and Human Services, and annual American Hospital Association (AHA) surveys of hospitals in 1982 and 1983. These national data show some differences between for-profit and not-for-profit hospitals.

OCR Data

The OCR survey sought data from all hospitals on admissions of uninsured patients during two weeks in 1981 (Rowland, 1984).3 Table 5.3 displays data on uninsured admissions as a proportion of total admissions in each type of hospital. Public hospitals accepted the greatest burden of uninsured patients—16.8 percent of their admissions—followed by not-for-profit hospitals (including teaching hospitals) with 7.9 percent, and for-profit hospitals with 6 percent.

TABLE 5.3

Inpatient Admissions by Source of Payment and Type of Hospital Ownership, United States, 1981 (millions of admissions).

Analysis of variations within ownership groups shows a consistent picture (Table 5.4). A slightly higher percentage of not-for-profit than for-profit hospitals (4.1 percent versus 3.1 percent) reported that more than 25 percent of the patients they admitted were uninsured. (The comparable number for public hospitals was 10.7 percent.) Conversely, a higher percentage of for-profit than not-for-profit hospitals (58.6 percent versus 44.5 percent) reported that less than 5 percent of the patients they admitted were uninsured. Since data are so often reduced to averages, these data on variability within categories are notable.

TABLE 5.4

Hospitals with a High or Low Volume of Uninsured Admissions, by Type of Ownership, United States, 1981.

On one measure, the OCR data showed no difference between for-profit and not-for-profit hospitals: 22 percent of emergency room visits were accounted for by uninsured patients in both types of hospitals. (The figure for public hospitals was 34 percent.) However, on another measure, for-profit hospitals saw relatively fewer emergency room patients than did not-for-profit hospitals: 1.4 versus 1.8 emergency room visits per hospital admission (Rowland, 1984).

In sum, for-profit and not-for-profit hospitals differ somewhat in the care of the vulnerable category of uninsured people, although the difference is relatively small. Furthermore, the fact that 6 percent of the patients admitted to for-profit hospitals are uninsured does not conform to the stereotype. Methodological questions arise because of the magnitude of the effort that was required of responding institutions and because of the limited (2-week) sampling period. Nevertheless, the OCR data are based on a very large sample (almost 5,800 hospitals) because they were collected as part of mandatory civil rights compliance efforts. No other source of national data exists on hospital services to uninsured patients, a plausible proxy for patients who are unable to pay for care.

AHA Data

The American Hospital Association's annual survey of hospitals includes information on bad debt and charity care as a percentage of hospital charges (i.e., the amount hospitals charge for the services provided over a period of time, in contrast to the amount they collect).4 The data are self-reported and are subject to bias because of low response rates, particularly on financial items. The 1982 data reported below are based on a 40 percent response rate; the nonresponse problem was greater among the for-profits than not-for-profits. The use of bad debt and charity dollars as a measure of uncompensated care itself presents problems. Although conceptually distinct and reported separately by hospitals, it is recognized that bad debt and charity care are neither generally nor consistently distinguished from each other for hospital accounting purposes. The distinction is probably made mostly by hospitals that have a positive reason to do so—for example, to demonstrate that they are meeting Hill-Burton "free care" obligations or to maximize reimbursement from those Blue Cross plans that include charity care (but not bad debt) as a reimbursable cost. Thus, although "bad debt and charity as a percentage of charges" reflects the vigor of debt-collection efforts, as well as willingness to serve patients who cannot pay, it is the most widely used measure of uncompensated care, and the AHA annual survey is the only national data source.

Data from the 1983 AHA survey (Table 5.5) show uncompensated care as 4.2 percent of the gross patient revenues in not-for-profit hospitals, but only 3.1 percent in for-profit hospitals (differences that are remarkably similar to the OCR data). (Both contrast sharply with public hospitals.) The 1982 AHA survey showed no clear difference between for-profits and not-for-profits (Table 5.5). In metropolitan areas, not-for-profit hospitals provided slightly more uncompensated care than did for-profit hospitals; the opposite was true in non-metropolitan areas. (No breakdown of independent and chain hospitals was available.) A regression analysis that included such variables as size, region, and teaching status, showed no statistically significant difference between not-for-profit and for-profit hospitals (Sloan et al. 1986); however, this analysis also controlled for "payer mix," a variable that appears to be closely associated with some types of uncompensated care (i.e., self-pay patients). Nevertheless, the lack of a clear difference between for-profit and not-for-profit hospitals in 1982 is apparent. Overall, the national data from AHA surveys provide weak support for the hypothesis that for-profit hospitals do less than not-for-profit hospitals to meet the needs of patients who are unable to pay.

TABLE 5.5

Uncompensated Care as a Percentage of Charges, by Ownership and Location, United States, 1982 and 1983.

State Data

Data from several states show a different picture. The committee sought data on uncompensated care in states where for-profit chain hospitals are an important presence. Data were obtained for five such states—California, Florida, Tennessee, Virginia, and Texas. The Texas data are from a special survey (response rate 80 percent) conducted by the Texas Hospital Association (1985) in connection with the activities of a state commission on the indigent care problem (Kent Stevens, Texas Hospital Association, personal communication, 1985). Data from the other four states are from state agencies to which hospitals are required to submit data. Except for the California data (which in-eluded only hospitals with 76-250 beds), these state data do not control for size, rural-urban differences, or teaching status. The data also do not indicate the presence or absence of nearby public providers—another factor that can influence the provision of uncompensated care in private hospitals.5

Table 5.6 shows bad debt and charity care as a percent of gross revenues in these five states. In California, which had a well-functioning system of public hospitals during the study years, the data show no difference between for-profit and not-for-profit hospitals. A different picture is seen in the other four states, where because of the characteristics of the Medicaid programs and the demographic makeup of the states, hospitals of all types provide higher levels of uncompensated care than is typical nationally. In Florida, Tennessee, Virginia, and Texas, for-profit hospitals have substantially lower bad debt and charity care deductions from gross revenues than do not-for-profit hospitals. Not-for-profit hospitals provide from 50 to over 150 percent more uncompensated care as a percentage of revenues in these states. Table 5.6 also shows that in the two states where data distinguish between chain and independent hospitals—Tennessee and California—the chain for-profit hospitals have lower charity care and bad debt rates than do independent for-profits.

TABLE 5.6

Hospital Uncompensated Care as Percentage of Gross Patient Revenues, Various States, 1981-1983.

These data and the data showing differences between metropolitan and nonmetropolitan areas indicate that location plays a role in the amount of uncompensated care that hospitals provide. Since poor people and nonwhite people are more likely than others to be uninsured and unable to pay for care, some would expect for-profit hospitals to avoid counties with relatively high poverty or nonwhite populations. A national study showed that this was not the case at the county level (Watt et al., 1986). More for-profit hospitals located in counties with slightly higher rates of poverty and nonwhite populations than did not-for-profit hospitals, but the differences were not statistically significant when controlled for census regions. On the other hand, for-profit hospitals—chain and independent—are more likely to be located outside of central cities than their not-for-profit counterparts, many of which of course made location decisions during an earlier period.

Reducing Uncompensated Care

Numerous strategies are available to both for-profit and not-for-profit hospitals that seek to minimize provision of uncompensated care. Transferring or "dumping" undesired patients has received considerable attention in the media. No available data indicate whether for-profit or not-for-profit hospitals are more likely to transfer patients for "economic reasons," or the extent to which health or lives are being endangered by such practices. However, there are concerns that in the face of a changing payment system and price competition, both for-profit and not-for-profit hospitals are increasingly ''dumping" unwanted patients. There are also fears that "dumping" will create serious financial stress on recipient hospitals. Chicago's Cook County, Hospital reportedly receives 6,000 inpatients per year from Chicago's private hospitals and an estimated 25,000-75,000 outpatients, all described as "dumped" (Schiff, 1985). One major public hospital, Parkland Memorial Hospital in Dallas, whose budget had been strained by transfers of out-of-county indigent patients for whom the hospital is not legally responsible, has instituted a "hot-line" to be used by referring hospitals. This is an attempt to ensure that medically unstable patients receive treatment before transfer and that "economic" reasons are not the only reason for transfer. Taking action to deflect uninsured patients is not unique to that hospital. In 1981 and 1982 alone, 15 percent of all hospitals adopted explicit limits on the amount of charity care they would provide. (Sloan et al., 1986).

Case studies conducted by the committee in three cities where for-profit chain, not-for-profit chain, independent not-for-profit, and public hospitals competed, showed that all types of hospitals have intentionally or unintentionally taken steps that can diminish their chances of providing indigent care (Townsend, 1986). The location of a hospital in relation to low-income populations and to other hospitals can heavily influence whether patients who are unable to pay will seek access, a factor that affected the siting decision of an investor-owned hospital in one of the cities. Other actions taken by for-profit, not-for-profit, or public hospitals in just these three cities include

• locating in neighborhoods with well-insured populations (for-profit and not-for-profit);

• having an emergency room that is not equipped for trauma, so that such cases (which produce disproportionate numbers of bad debts) are taken elsewhere by ambulance drivers (for-profit);

• refusing admission of uninsured patients and referring them to a public hospital, sometimes as much as two hours away (this strategy was sometimes defeated by uninsured pregnant women, who would wait in the parking lot until they were in late stages of labor before entering the emergency room) (for-profit and not-for-profit);

• deciding not to provide (for-profit) or to stop providing (not-for-profit) obstetric services—a service that often produces disproportionate numbers of bad debts;

• screening for financial status before admitting and admitting only urgent cases, which are then transferred to a public hospital after stabilization (for-profit and not-for-profit);

• requiring preadmission deposits (for-profit and not-for-profit).

The case studies also showed that the most dramatic action taken to reduce uncompensated care was the public hospital that dosed its emergency room so it could shift uninsured patients to two nearby religiously affiliated hospitals. This example shows that hospitals of all types may act to reduce the amount of uncompensated care that they provide.

This brief list of strategies used by hospitals to deflect nonpaying patients suggests that it is difficult to avoid providing some uncompensated care. Case law, some state statutes, and accreditation standards all hold hospitals responsible for providing services in medical emergencies (although it is often not clear what constitutes an emergency or what services are required). The case studies showed several other reasons why hospitals would accept some level of uncompensated care as the price of doing business. Maintenance of good medical staff relations may sometimes require hospitals to allow physicians to admit a few patients who lack means to pay. Also, some services that attract indigent patients may also draw enough paying patients to result in a net revenue gain. Moreover, it may not be possible to reduce bad debt below a certain minimum in any service industry that cannot operate on a cash-and-carry basis. Data on bad debts do not by themselves reveal how assiduously institutions tried to minimize the provision of uncompensated care.

The Impact of For-Profit Acquisition

The acquisition (and construction) activities of investor-owned hospital companies could, in theory, have both positive and negative impacts on access to care. The positive hypothesis, largely undocumented, is that their investments may at minimum make services more convenient to the people they serve. The growth of investor-owned hospitals in areas of relatively high population growth supports this idea, their relatively low hospital occupancy rates notwithstanding. Anecdotes suggest that some acquisitions may have prevented the closure of hospitals, a notion supported by data showing that hospitals purchased by for-profit chains in California were unprofitable prior to acquisition (Pattison, 1986). However, in few cases are for-profit hospitals the only source of care available to populations; fewer than a dozen of the 365 hospitals on the Health Care Financing Administration's list of "sole community hospitals" are for-profit.6

The negative hypothesis is that acquisitions by investor-owned companies will reduce the amount of services provided to patients who are unable to pay. Two studies conducted for the committee, which examined data on small numbers of hospitals before and after acquisition,7 provide some support for this concern. One study describes for-profit chain acquisitions between 1979 and 1981 in Florida; the other examines acquisitions in California between 1977 and 1981 (Brown and Klosterman, 1986; Pattison, 1986). Because most acquired hospitals were previously for-profit, the changes observed cannot be attributed to a change from not-for-profit to for-profit ownership but, rather, to changes in goals and strategies.8

At hospitals purchased by investor-owned corporations in Florida, the percentage of total patient revenues for charity care and bad debt declined between 14 and 35 percent in three years, while hospitals that had not changed ownership showed an average 5 percent increase in the same measure of uncompensated care (Brown and Klosterman, 1986). In California, hospitals acquired by for-profit chains reportedly reduced bad debt from 2.7 percent to 0.2 percent of charges within four years of acquisition (Pattison, 1986). At least some of the reduction in uncompensated care may result from initiation of more effective collection procedures by the new owners.

This evidence on reduced uncompensated care after hospital acquisitions by investor-owned companies is based on a small number of cases and a problematic measure (bad debt and charity care as a percentage of gross patient revenues). Yet, there may be reason for concern, particularly because acquisitions in the 1980s began to include more public and not-for-profit hospitals than in earlier years (Hoy and Gray, 1986). Some protective mechanisms are available for communities that fear a reduction in indigent care after an investor-owned purchase of a hospital, some protective mechanisms are available, particularly if a local government hospital is bought. For instance, the money paid for the hospital may be placed in a fund devoted to payment for indigent care, the purchase agreement may require provision of some amount of charity care, or a buy-back clause may be inserted in the purchase agreement that will enable the local authority to regain control of the hospital if it is not satisfied with the administration of the facility.

Cross-Subsidies and Uncompensated Care

Although some hospitals have significant nonpatient care revenue sources (Table 5.2), revenues from paying patients are key to the economic health of many institutions that provide substantial amounts of uncompensated care. Hadley et al. (1982) found that in 1980, one-third of the hospitals providing a high volume of care to poor people (i.e., with more than 24 percent of charges going to Medicaid, charity care, and bad debt) were financially "stressed," having deficits on operating and total accounts. Among hospitals that provided high amounts of uncompensated care, the main factor imposing financial stress was a relative lack of revenues from charge-paying, commercially insured patients from which to subsidize uncompensated care.

A frequently stated concern is that the success of for-profit hospitals in attracting paving patients could erode the ability of other hospitals in their communities to subsidize indigent care. However, it is obvious that any hospital that attracts paying patients and serves few indigent patients could have this effect on hospitals that attempt to cross-subsidize indigent care.9 Although these problems and concerns are very real, the committee found no systematic data on the impact of investor-owned hospitals on other hospitals that cross-subsidize uncompensated care.10

Other For-Profit Providers

Perhaps a more serious threat to the ability of institutions to cross-subsidize uncompensated care is in the growth of freestanding alternative sites that provide such services as urgent care, certain surgical procedures, radiological procedures, and the like. These centers tend to be for-profit, whether they are owned by physicians, investor-owned corporations, or not-for-profit hospitals. The services are usually designed to offer more convenient locations and hours, and lower charges than those found at traditional sites, such as hospitals and physicians' offices. Thus, they undoubtedly enhance access in some respects. However, although systematic data are not available, it appears that many—perhaps most—of the new alternative sites provide little uncompensated care; typically, cash, credit card, or evidence of workmen's compensation eligibility is demanded at the time of service.

Such freestanding providers attract paying patients needing those services on which many hospitals have generated surpluses that could be used to make up losses on uncompensated care. Thus, although the new providers may improve access to care for certain segments of the population, the segment that experiences major financial barriers—the poor and uninsured—could be hurt if hospitals respond to the revenue loss by reducing uncompensated care. While systematic documentation is lacking on this effect, the emergence of freestanding providers is clearly one of several factors making it more difficult for hospitals to cross-subsidize uncompensated care.

Types of Services in Various Hospitals

It is widely agreed that hospitals lose money on some types of services either because of difficulty in charging patients for full costs (services that are infrequently used may present this problem) or because the service attracts an unusually large proportion of uninsured patients. It has frequently been alleged that for-profit institutions are more likely than not-for-profits to confine their operations to profitable services, which either deprives the community of access to certain services or throws an extra burden on institutions that do offer the services. Thus, the committee sought information on which services are unprofitable for institutions and which services for-profit institutions tend to offer or not to offer. Data are more readily available on the latter question.

Which Hospital Services Lose Money?

Unfortunately, only scattered and unsatisfactory, data are available in answer to this question. Several types of services have been identified for which at least some hospitals are less likely to receive full payment. In one tertiary care institution, services that utilize low- or mid-level technology were more likely to be uncompensated than were such procedures as coronary bypass, hip replacement, and peripheral vascular surgery, perhaps because it was both feasible and important to obtain assurance of payment before such elective surgery was performed. National Discharge Survey data show that maternity and accident cases are heavily represented among "self-pay" patients, which are a primary source of bad debts (Sloan et al., 1986). Estimates derived from the Census Bureau's 1984 Current Population Survey show that 25 percent of women in the prime childbearing years of 18-24—when 40 percent of births occur—had no health insurance (Gold and Kenney, 1985). At Vanderbilt Hospital, a regional neonatal care center, the treatment of newborns accounted for 27 percent of the entire institution's uncompensated charges (Sloan et al., 1986). More inferential evidence comes from regression analysis of national data, which show that margins per case (revenues less expenses) for an entire institution are negatively related to the volume of births therein (Watt et al., 1986).

Such data suggest why obstetrical services (along with emergency rooms, to which come trauma victims, who are often uninsured are often identified as services that generate disproportionate amounts of uncompensated care. This is undoubtedly true at some institutions, but not all of them. Depending on a variety of factors having to do with controlling bad debt and with stimulating other services (emergency rooms being a major source of admissions), obstetrical services and emergency departments can contribute to an institution's bottom line. For example, analyses of hospital operating statements suggest that public hospitals and large not-for-profit teaching hospitals in major metropolitan areas suffer large outpatient clinic losses and often admit unfinanced patients for essential inpatient services. In contrast, not-for-profit and investor-owned hospitals in economically advantaged areas usually earn surpluses on outpatient diagnostic and treatment services. Similarly, whether obstetrics is a money-losing service depends on such factors as institutional location and admitting policies. It is significant that in some cities, well-publicized amenities (champagne and gourmet meals) are used to attract obstetrical patients (well-financed ones, presumably) and that one of the most profitable of the investor-owned hospital companies, Humana, Inc., owns several women's hospitals. Furthermore, with the trend toward vertical integration and institutional marketing of a complete array of services, there may be serious competitive disadvantages in being unable to provide basic services.

In sum, the committee was unable to find satisfactory information about how often (and under what circumstances) it is a net financial drain on an institution to offer particular services. However, although obstetrics and emergency services may sometimes provide economic rewards to an institution, it is known that these services (or at least some aspects of these services—e.g., having an emergency room open in the early morning hours) are often money losers.

For-Profit/Not-For-Profit Differences in Services

At the committee's request the American Hospital Association's Hospital Data Center provided 1983 data showing how the facilities and services reported on the AHA's annual survey were distributed in hospitals of different size and ownership types. The validity of such data are often questioned, because hospitals may exaggerate their services and because a report that a hospital has a given service reveals nothing about how often it is used. Nevertheless, the AHA is the only source of national data on what services and facilities hospitals offer. The 1983 data appear in Table 5.A. 1, which is appended to this chapter.

With data presented on 42 services, 5 sizes of hospitals, and 5 types of ownership, the results of the comparisons are not easily described. However, several points can be made regarding the services offered by investor-owned chain hospitals.

First, focusing on the modal-size category (100-199 beds) and larger institutions, there is a set of basic services that virtually all hospitals offer and that differ little between investor-owned chain hospitals and not-for-profit hospitals—emergency rooms, post-operative recovery rooms, respiratory therapy, physical therapy, pharmacies. Most investor-owned chain and not-for-profit hospitals also have blood banks, histopathology laboratories, and diagnostic radiation services.

However, there is a large group of services that are more commonly offered in not-for-profit hospitals (chain and independent) than in investor-owned chain hospitals and very few that are more common in investor-owned chain than not-for-profit hospitals. Not-for-profits are more likely to have an outpatient department, premature nursery, dental services, hospice care, home care, hospital auxiliary, health promotion services, family planning services, various types of psychiatric services, radiation therapy, radioisotope implants, and therapeutic radioisotopes. (Some of these differences may change, in the view of some, because changing payment incentives may stimulate the availability of outpatient and home health services among all types of hospitals.) More commonly available in investor-owned chain hospitals are abortion services and patient representatives; among smaller hospitals, investor-owned facilities are more likely than not-for-profit institutions to offer podiatric services, ultrasound, CT scanners, and diagnostic radioisotopes.

The data also show a pattern of more services being offered by investor-owned chain hospitals than by independent proprietary hospitals, suggesting that the growth of the investor-owned hospital companies may have increased the availability of a broader range of services at the hospitals they acquired.

In sum, the AHA data provide some evidence of a narrower range of services in investor-owned hospitals, compared with not-for-profit hospitals. However, the data do not permit judgment on either of two crucial items. First, although some of the services that are found less frequently in for-profit hospitals seem unlikely to be money-makers (e.g., health promotion services, hospital auxiliaries), the extent to which the greater breadth of services provided by not-for-profit institutions includes money-losing services that must be cross-subsidized from other services is not clear. Second, the data do not show the extent to which services not offered by hospitals are essential (or even important) services that are not otherwise available in the community.

Regarding the narrower topic of emergency services and obstetrical services, few inferences can be drawn. Almost all hospitals in the AHA survey reportedly having emergency services, and the difference between investor-owned and not-for-profit hospitals is only a few percentage points. (Differences in having outpatient departments were slightly larger.) A study of Florida nonteaching hospitals with fewer than 400 beds showed that investor-owned hospitals were slightly more likely than not-for-profit hospitals to have an emergency room (80 percent versus 72.5 percent) (Sloan and Vraciu, 1983). It is likely that institutional variations in what their emergency services actually consist of and what screening procedures are used before services are provided are more important than the presence or absence of an emergency room in determining the extent to which a facility provides access, particularly to people who are unable to pay.

The committee found no sources of data on for-profit/not-for-profit differences in hospitals having obstetrical services.11 But national data show that for-profit hospitals (both chain and independent) have significantly fewer births as a percentage of admissions than do not-for-profit hospitals (Watt et al., 1986). The AHA data, as well as data from Florida, which show that premature nurseries are much less common in investor-owned than in not-for-profit hospitals (Sloan and Vraciu, 1983), are also suggestive.12 However, no means are available to determine whether this pattern reflects a strategy to minimize uncompensated care or the decision (by hospitals or planning agencies) not to offer services that are already being amply met in communities.

Cross-Subsidies and Uncompensated Care

Still another access-related concern about the growth of investor-owned hospital companies is whether they may be less willing than not-for-profit organizations to sustain a money-losing hospital or one that fails to attain economic goals. Although the closure of a hospital in a region with substantial excess capacity is not necessarily a tragedy, the closure of hospitals on which large numbers of uninsured patients depend or which are the only source of care in a given locale could jeopardize access to care.

Because the management of investor-owned companies is responsible to shareholders to make profitable use of invested capital, and because not-for-profit hospitals have a long history of losing money on operations, a differential willingness to close money-losing facilities may seem self-evident. However, the survival of a not-for-profit (or public) hospital that persistently loses money on patient care services depends on its ability to obtain nonpatient revenues. Clearly, some are more successful than others. Nor is it evident that an investor-owned company would readily walk away from a money-losing facility in which it had made a substantial investment without carefully evaluating the options that might be available. In short, the topic of hospital closure and type of ownership is a complex empirical (as well as theoretical) issue.

Studies of hospital closures show that for-profit hospitals are disproportionately likely to close (Sloan et al., 1986; Mullner et al., 1982). However, the long historical decline in the number of independent, proprietary for-profit hospitals and the growth of investor-owned chains during the past 15-20 years suggest that it is the former that have been closing, not the latter. To assess this possibility, the committee obtained data on all 540 hospitals that had ever been acquired or constructed by the six largest investor-owned companies13 (which owned two-thirds of all investor-owned hospitals in 1984) (Hoy and Gray, 1986). Of the 540 hospitals, 12 (2.2 percent) had been closed (a number that does not include hospitals that were replaced). In most cases, the closure of a hospital was associated with the addition of beds at another facility, the construction of a new facility, that replaced two others, or the conversion of a facility into a different type of facility. Very few hospitals were simply closed.14

In short, to date investor-owned hospitals have not shown a propensity, to dose. Indeed, in view of the likelihood that some of the hospitals they have acquired had been in serious financial trouble (Pattison, 1986), and in view of the overall low occupancy rates among investor-owned hospitals (Peter Kralovec, Hospital Data Center, American Hospital Association, unpublished data, 1985) a more plausible criticism, ironically, is that they have contributed to the nation's surplus bed capacity.

The available information pertains to a time when most hospitals experienced relatively little trouble in maintaining economic viability. If increasing competition, declining occupancy rates, new governmental regulations, or a major economic recession erode hospital margins generally, it becomes increasingly likely that some hospitals of all ownership types will become financial liabilities. Some observers predict the closure of as many as 20 percent of the nation's hospital beds in the next 5-10 years. More serious (and plausible) than concerns about investor-owned hospitals closing is the likelihood that some of the hospitals that are the most vulnerable to closure are the ones that are performing a vital role in serving uninsured populations. If that is the case, as studies of financially distressed hospitals suggest, strong public policies are needed to mitigate the growing crisis created by an uninsured population of almost 35 million people in an increasingly competitive health care system.

Access to Nursing Homes

Whereas the major concerns regarding hospital care pertain to uninsured patients, a major access concern of nursing homes involves low-income patients covered by the federal/state Medicaid program. Between people who are Medicaid-eligible when they enter a nursing home and "private-pay" patients who deplete their resources until they become eligible for Medicaid, the Medicaid program has become the largest source of payment for nursing home care. However, because Medicaid payment rates are often substantially lower than the rates other patients pay, nursing homes tend to prefer private-pay patients. The incentive to select non-Medicaid patients is illustrated by 1983-1984 data showing that nursing home return-on-equity in California declined as Medicaid utilization increased (Hawes and Phillips, 1986). A second major access issue involves "heavy-care" patients (e.g., with multiple medical problems) for whom payment (e.g., under Medicaid) is set on a per-diem basis, but whose care is costly.15

Nursing home operators often have the option of being selective in admissions and of charging private-pay patients more than Medicaid patients. This is because the demand for nursing home beds greatly exceeds the supply in many states and because adequate alternatives do not often exist for people in need of long-term care or assistance. Thus, nursing homes usually have high occupancy rates and often have waiting lists, enabling operators to select private-pay or "light-care" individuals over less-remunerative patients. Methods used by for-profit and not-for-profit nursing homes to select the most financially advantageous patients include separate waiting lists for private-pay and Medicaid patients, discharging private-pay patients to hospitals when their financial resources are gone, holding vacant beds open until a private-pay patient becomes available, requiring large pre-admission "contributions," and selecting "light-care" Medicaid patients certified to be in need of skilled nursing care.16

Regarding ownership and access issues in nursing homes, among which for-profit ownership predominates, for-profit homes appear to serve a disproportionately large number of Medicaid patients. Although studies at the local and state levels are somewhat inconsistent,17 the most recent national data, from the 1977 National Nursing Home Survey, show that the percentage of Medicaid patients in for-profit nursing homes was substantially larger than in not-for-profit homes (Table 5.7). (The data do not differentiate between chain and independent homes.) Although economic incentives can discourage admission of "heavy-care" patients, and although such patients do not always receive the specialized care they need, the committee found no studies that compare the behavior of for-profit and not-for-profit nursing homes with regard to "heavy-care" patients.

TABLE 5.7

Payment Source for Nursing Home Residents by Type of Facility Ownership, 1977 (number and percent of residents).

Issues and Recommendations

The Measurement of Care to Patients Unable to Pay

"Uncompensated care" (deductions from gross revenues for bad debt and charity care) is a seriously flawed measure of either institutional performance or the extent to which the needs of those who are unable to pay are being met. (The number of uninsured patients served is a better measure, but it is less widely available.) To say that a hospital has a given percentage of bad debt does not reveal precisely whether it has been acting with generosity or whether it has been lax or ineffective in trying to collect payment. Furthermore, "uncompensated care" is not a measure of an institution's real costs in providing such care, but only of what revenues would have been gained if payment had been received. Finally, expressed as a percentage of gross patient revenues, "uncompensated care" does not reflect any nonpatient care revenues that may be obtained to subsidize uncompensated care. Nevertheless, uncompensated care as a percentage of gross revenues is the most commonly used measure of institutions' service to patients who are unable to pay. It is useful for comparisons across categories, but it should not be taken as a true measure of the extent to which human needs are being met. And, because not all persons seek needed care, uncompensated care (or number of uninsured patients served) is at best a partial proxy for the full unfinanced needs of the population.

The Need for Better Information

Clearly, there is a need for better information about the extent to which hospitals and other health care institutions are admitting and treating patients who are unable to pay or are providing services that are unprofitable, about changes in the provision of uncompensated care, and about why these changes are occurring. The committee's ability to draw conclusions about the effect of investor-owned providers on access to care has been limited by the nature of available data. However, much of the data that has been summarized in this chapter can provide a baseline against which changes among providers of all types can be compared.

Accordingly, the committee recommends the monitoring and study of the following topics and issues:

• The number of people unable to pay for care who (1) needed care but did not try. to obtain it and (2) tried to obtain care but were refused for financial reasons.

• The amount of uncompensated care, both absolutely and relative to gross revenues, provided by hospitals of different ownership types. This will enable policy-makers and others to monitor changes across all types of hospitals and differences among types of hospitals, which may change over time. Controlling for size of hospital, location, and teaching status will help ensure that factors known to affect uncompensated care are accounted for. Breaking out data by whether hospitals are sole providers or are located near a public hospital will help in assessing the impact on access of differences among hospitals in the amount of uncompensated care provided.

• The types of services and the amount of each type of service provided by hospitals of different ownership. Local-level studies and controls on whether alternative providers of care will help show whether absence of a service in a particular hospital represents deprivation for a community.

• The provision of care to the poor and medically indigent in other service settings, such as nursing homes, ambulatory surgery centers, dialysis centers, birthing centers, and outpatient diagnostic centers.

• Before-and-after studies of acquisitions of hospitals by for-profit and not-for-profit hospital systems, to examine changes in the provision of uncompensated care and the number and types of services provided.

• An evaluation of the effects of state and local governmental programs that were designed to increase the resources available to those unable to pay for care. These studies should examine the extent to which differences in the mix of hospital ownership (state by state) affect the kinds of programs adopted, the ways in which the programs are implemented, and their eventual effects.

Such information will be of growing importance as competitive conditions make it more difficult for institutions to finance important activities. State health planning agencies, and the 141 local health planning agencies that still exist, can play an important role in obtaining and publicizing information about problems of access to health care.18

The Obligations of Health Care Institutions

What are the social, moral, or legal obligations of hospitals and other health care organizations to serve people who need care, but who lack the ability to pay? From whence might such an obligation derive, and what is its extent (e.g., to emergency situations only or to more than that)? A full analysis of such questions is beyond the scope of this study. However, it is important to keep in mind why the questions arise and to consider their implications for the for-profit/not-for-profit distinction.

The questions arise partly because the care of millions of uninsured and underinsured people may depend on the willingness of institutions to provide services for which they may not get paid. The question of institutional obligation to provide care to those who cannot pay (or to provide services that lose money) is, thus, also a question of whether institutions are obligated to raise funds for this purpose. Historically, many health care institutions were given or accepted such an obligation, and it has been part of the traditional ethos of many hospitals, particularly not-for-profit and public institutions. However, these institutions also had access to nonpatient revenues—governmental grants and appropriations, charitable contributions, and the like. The availability of these crucially important sources of revenue has dramatically diminished in proportion to need,19 and institutions have long had to subsidize the care of some patients with revenues derived from charge-paying patients. Their ability to do this depends on the number of such patients that they serve and the impunity with which they can raise prices for the services they provide.

The committee, like the President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research (1983) and many other bodies before it, believes that ensuring an adequate level of health care is a societal obligation (for which provisions can be made in both the public and private sectors) and that government should, and eventually must, make provisions to assure care for those whom the private sector fails.

However, the committee also believes that government's failure to make such provisions does not relieve health care institutions of their moral obligation to help care for those who are unable to pay. The committee believes that in the absence of full governmental funding of health care for all those who are unable to pay, health care institutions should be expected to do whatever they can to care for needy members of their communities. At minimum, health care institutions, because of the peculiar vulnerability of people who need medical care but lack resources to pay and because of the traditional values that have been attached to providing health care, are morally obligated—and should be legally obligated—to provide medical care in situations of emergency. (The definition of emergency and the extent of care that must be provided are slippery issues. See, for example, Relman, 1985; Annas, 1985.) Beyond this, the amount of such care that should be provided depends on various factors, including need in the community and the ability of the institution to provide uncompensated care without imprudent financial strain.

The Issue of Tax Status

A question that arises is whether taxpaying and tax-exempt institutions have the same or different moral or legal obligations to provide uncompensated care in other than emergency situations. Although the committee had insufficient discussion on which to base strong recommendations, the matter of taxes and tax exemptions merits much additional consideration and discussion in health care. Several issues are discussed below.

The Question of "Fair Share." Our "system" of health care for the uninsured has consisted of undefined, unorganized, and uncoordinated institutional and professional contributions, financed largely by the shifting of costs onto insured patients. It is, accordingly, difficult even in theory to provide a standard for determining whether a particular hospital has clone its "fair share" of charity care or, indeed, whether the notion of "fair share" makes sense (Brock and Buchanan, 1986). We can examine comparative performance data, as this report has, but these data alone do not lead to conclusions about whether hospitals have satisfied their obligations, if indeed such obligations exist. All we know for certain is that without governmental assistance, many people will not get adequate medical care unless enough hospitals provide a substantial amount of uncompensated care.

Obligations Attached to Tax Exemptions. In the committee's view some social obligations or expectations should be attached to an institution's tax-exempt status. At the local level, as the Supreme Court of the State of Utah recently held, it is legitimate for local governments—recognizing the extent to which exemptions from property taxes entail a taxpayer subsidy of a health care institution—to evaluate what benefits are received in exchange and to act accordingly (Utah County v. Intermountain Health Care, Inc., 1985). The more health care institutions behave as a commercial enterprise might be expected to behave—for example, in not obtaining or dispensing charity care—the more problematic is the justification for property, tax exemptions. Similar reasoning can be applied to federal exemptions from corporate income taxes. However, it should be remembered that the social obligation of a not-for-profit institution can be discharged in many ways (which includes offering unprofitable services and engaging in inadequately funded educational activities and unsponsored research); providing significant amounts of uncompensated care is but one.

Obligations of For-profit Institutions. Do for-profit hospitals have a lesser responsibility to provide uncompensated care than other kinds of hospitals? The answer depends on values and points of view. From the standpoint of an indigent patient in need of care, but ineligible for publicly funded care, an argument that the local hospital has paid its taxes and thereby has discharged its social obligations would clearly be of little comfort.

From the standpoint of a taxpaying health care organization, the argument that its taxes at least partly discharge its social obligations is more telling. Since the taxpaying (i.e., for-profit) institution is not afforded the benefits of a tax exemption, the argument goes, it should not be saddled with the obligations that attach thereto.

Empirically, the question of whether for-profits accept as great a burden in taxes and charity care as not-for-profits accept in charity care alone is not answered simply. The tax data needed to make an exact comparison are not easily available, but a rough estimate can be made. In 1983 the four largest investor-owned multihospital companies (Hospital Corporation of America, Humana, National Medical Enterprises, and American Medical International) paid on average 2.5 percent of gross revenues in income taxes20 (Steven C. Renn, The Center for Hospital Finance and Management, The Johns Hopkins University, personal communication, 1985). If their level of uncompensated care is reflected in the for-profit average reported by the AHA for 1983—3.1 percent of gross revenues, as shown in Table 5.6—the sum of income taxes and uncompensated care (5.6 percent of gross revenues) exceeded the 4.1 percent of gross revenues that not-for-profit hospitals accounted for as uncompensated care. Accepting the argument that taxes are in some sense equivalent to uncompensated care, and making the somewhat problematic assumption that not-for-profit hospitals are doing their fair share, these four companies together can claim to have done their ''fair share," if such average figures are meaningful given the highly variable nature of local uncompensated care problems and of the resources available to meet them.

In sharp contrast is the point of view that payment of taxes is irrelevant to a health care institution's purpose and obligations. From this viewpoint, social policy should insist that hospitals be public service institutions. Whether they choose to operate as not-for-profit or for-profit enterprises, hospitals should be made to serve public needs. There is no fundamental right to public support in the operation of a hospital for private ends—whether those ends are to generate profits, create employment, or engage in educational and research activities. From this point of view, the public can, and should, set some conditions to guarantee access to the services it needs. Community support, whether financial or otherwise, should be contingent on performance in the public interest. This includes, of necessity, some activities that are not profit maximizing. Whether those who wish to operate the hospitals under these conditions want to organize themselves on a for-profit or a not-for-profit basis is a matter of secondary interest. If hospitals can make a profit while performing their functions to the community's satisfaction, however that might be expressed, everyone comes out ahead; but there is no special reason to excuse any such institution from public service.

Although the conflicting positions just discussed cannot be fully reconciled, the committee holds that all health care institutions have social obligations that flow from the needs of the communities that they serve and the ethical traditions of health care. However, taxing authorities that grant tax exemptions are entitled to expect not-for-profit organizations to continue, within reasonable limits, to provide some form of service in exchange for the tax exemptions, whether that service be providing uncompensated care, offering unprofitable services, sponsoring educational or research activities, or offering care at reduced prices. The feasibility of institutions' generating the revenues required to provide all needed care to uninsured and underinsured patients is becoming increasingly doubtful.

The Government's Obligation for Uninsured Patients

Notwithstanding the committee's belief that institutions have an obligation to do whatever they can to meet the health care needs of people who are unable to pay, it is increasingly clear that new ways of providing financial support and encouragement for providing uncompensated care are needed.

Cross-subsidization at the institutional level has never been a satisfactory or adequate method of meeting the health care needs of uninsured people. People often do not obtain needed care, and institutions that provide large amounts of uncompensated care tend to become financially distressed (Hadley et al., 1982; Kelly and O'Brien, 1983). Furthermore, it is apparent that institutions will find it increasingly difficult to cross-subsidize indigent care. Providers are under increasing pressure from third-party payers to compete on price. Institutions in such circumstances—probably only a small number today, but likely many more in the future—risk their very survival if they engage in behavior that makes them noncompetitive on price. This being the case, in the absence of governmental intervention, the likelihood is that care for those who are unable to pay will be increasingly jeopardized.

The committee did find some evidence that hospitals of all types are attempting to diminish their uncompensated care burden. Although it is likely that some hospitals are decreasing their uncompensated care load because of financial necessity, the committee is concerned that others are reacting not to financial stress, but rather to a change in the ethos of health care. Maturation of the health care market has made growth more difficult; efforts by purchasers of care to control costs are beginning to make competition a way of life; and health care increasingly is becoming commercialized. These developments may be eroding the charitable attitude that has characterized many hospitals. Notions of responsibility for community health needs and acceptance of charitable activities as an inherent part of the provision of health care services may be disappearing. These changes in the spirit of health care may allow uncompensated care burdens to be reduced without fear of public or professional opprobrium.

Federal action to solve access problems for the poor is not imminent, but there is an encouraging and growing movement among state and local governments to enhance access to care for people who lack a source of funds. A variety of financing mechanisms are possible;21 evaluation of efforts now under way at the state level is needed. How states choose to approach the problems of financing care for those who are unable to pay will depend on political circumstances, and the access problems will have to be identified. In the committee's view, providing medical care to the nation's 35 million uninsured people is a challenge that should be very high on the nation's public policy agenda.

Conclusion

The major access issue in the United States concerns patients who are unable to pay for care and who are dependent on the willingness of hospitals to provide services. Two measures of such willingness were examined in this chapter: admissions of uninsured patients to hospitals and provision of uncompensated care by hospitals. The performance of not-for-profit hospitals was more favorable on both measures, although when measured as percentages of total admissions or total revenues, the national differences were not large. Small percentage differences, however, can translate into large numbers of patients, particularly if institutions that provide comparatively small amounts of uncompensated care comprise a relatively large proportion of the market. Data from four of five states about which the committee obtained data showed that not-for-profit hospitals provided two or three times as much uncompensated care, on average, than did for-profit hospitals. (Both types provided less such care than did public hospitals.) Because revenues from paying patients are key to the ability of many institutions to provide uncompensated care, and because the amount of revenue needed depends on the amount of uncompensated care provided, the for-profit presence in such circumstances may make it more difficult for other hospitals to provide uncompensated care.

Freestanding ambulatory care centers (which tend to be for-profit) have a similar effect. However, little direct evidence is available on the question of how some institutions impact on other institutions. The question of whether for-profit hospitals eschew nonprofitable services also could not be answered satisfactorily. Although larger percentages of many services are offered in not-for-profit than for-profit hospitals of similar size, satisfactory evidence is not available on which services are unprofitable.

In the view of the committee, traditional values of health care institutions remain important, meaning that health care institutions should do whatever they can to meet the needs of uninsured patients. The committee also concluded that tax-exempt institutions should be reasonably expected to accept a heavier responsibility for actions that depart from profit-maximizing behavior.

Finally, the committee concluded that the problem of access for uninsured patients cannot be dealt with adequately by health care institutions, and that the expectation that they will do so is a major public policy failure. A variety of options are available to address this problem, but as our health care system becomes more competitive and price sensitive, the resulting impact on uninsured patients is a major problem that should no longer be ignored.

References

• American Association of Fund Raising Counsel, Inc. (AAFRC) (1985) Giving USA: A Compilation of Facts and Trends on American Philanthropy for the Year 1984. New York: American Association of Fund Raising Counsel, Inc.

• American Hospital Association (1984) Hospital Statistics. Chicago, Ill.: American Hospital Association.

• American Hospital Association (1985) Economic Trends 1.

• Annas, George J. (1985) Adam Smith in the Emergency Room. Hastings Center Report 15(August):16-18. [PubMed: 4055350]

• Bernstein, Arthur H. (1984) Guidelines Issued for Preserving Tax-exempt Status. Hospitals 58(May 1):102. [PubMed: 6706340]

• Brock, Dan W., and Allen Buchanan (1986) Ethical Issues in For-Profit Health Care. This volume.

• Brooks, Charles H., and John H. Hoffman (1978) Type of Ownership and Medicaid Use of Nursing Care Beds. Journal of Community Health 3(Spring):236-244. [PubMed: 363752]

• Brown, Kathryn J., and Richard E. Klosterman (1986) Hospital Acquisitions and Their Effects: Florida, 1979-1982. This volume.

• Farley, Pamela J. (1985. a) Private Insurance and Public Programs: Coverage of Health Services. Data Preview 20, National Health Care Expenditures Study. Rockville, Md.: National Center for Health Services Research and Health Care Technology,.

• Farley, Pamela J. (1985. b) Who Are the Under-insured? Milbank Memorial Fund Quarterly 63(Summer):476-503. [PubMed: 3934581]

• Feder, Judith, Jack Hadley, and Ross Mullher (1983) Poor People and Poor Hospitals: Implications for Public Policy. Working Paper 3179-06. Washington, D.C.: Urban Institute.

• Federation of American Hospitals (no date) Statistical Profile of the Investor-Owned Hospital Industry 1983. Washington, D.C.: Federation of American Hospitals.

• Gold, Rachel B., and Asta M. Kenney (1985) Paying for Maternity Care. Family Planning Perspectives 17(May/June):103-110. [PubMed: 3916182]

• Hadley, Jack, Ross M. Mullner, and Judith Feder (1982) The Financially Distressed Hospital. The New England Journal of Medicine 307(November 11):1283-1287. [PubMed: 6752711]

• Hawes, Catherine, and Charles D. Phillips (1986) The Changing Structure of the Nursing Home Industry and the Impact of Ownership on Quality, Cost, and Access. This volume.

• Hoy, Elizabeth, and Bradford H. Gray (1986) Trends in the Growth of the Major Investor-Owned Hospital Companies. This volume.

• Kelly, Joyce v., and John J. O'Brien (1983) Characteristics of Financially Distressed Hospitals Cost and Utilization Project. Research Note 3. Rockville, Md.: National Center for Health Services Research.

• Mullher, Ross M., Calvin S. Byre, Paul Levy, and Joseph D. Kubal (1982) Closure Among U.S. Community Hospitals, 1976-1980: A Descriptive and a Predictive Model. Medical Care 20(July):699-709. [PubMed: 7121089]

• National Center for Health Statistics (1979) The National Nursing Home Survey: 1977 Summary for the United States. Vital and Health Statistics, Series 13, No. 43. Washington, D.C.: U.S. Government Printing Office. [PubMed: 506073]

• Pattison, Robert V. (1986) Response to Financial Incentives Among Investor-Owned and Not-for-Profit Hospitals: An Analysis Based on California Data, 1978-1982. This volume.

• President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research (1983) Securing Access to Health Care. Vol. 1: Report. Washington, D.C.: U.S. Government Printing Office.

• Relman, Arnold S. (1985) Economic Considerations in Emergency Care: What Are Hospitals For? The New England Journal of Medicine 312(February 7) :372-373. [PubMed: 3969089]

• Robert Wood Johnson Foundation (1983) Updated Report on Access to Health Care for the American People. Princeton, N.J.: The Robert Wood Johnson Foundation.

• Rowland, Diane (1984) Hospital Care for the Uninsured: An Analysis of the Role of Proprietary Hospitals. Paper prepared for the Annual Meeting of the American Public Health Association, Anaheim, California.

• Schiff, Gordon (1985) Letter to the Editor. The New England Journal of Medicine 312(June 6):1522.

• Sloan, Frank A., Joseph Valvona, and Ross Mullner (1986) Identifying the Issues: A Statistical Profile. In Frank A. Sloan, editor; , James F. Blumstein, editor; , and James M. Perrin, editor. (eds.) Uncompensated Hospital Care: Rights and Responsibilities. Baltimore, Md.: Johns Hopkins University Press.

• Sloan, Frank A., and Robert A. Vraciu (1983) Investor-Owned and Not-For-Profit Hospitals: Addressing Some Issues. Health Affairs 2(Spring):25-37.

• Texas Hospital Association (1985) Table from a THA Survey of Uncompensated Care in Hospitals, published with "THA Statement of Fair Share Formula for Financing Care for the Medically Indigent." .

• Townsend, Jessica (1986) Hospitals and Their Communities: A Report of Three Case Studies. This volume.

• U.S. Bureau of the Census (1985) Economic Characteristics of Households in the United States: Fourth Quarter 1983. Current Population Reports, P-70-83-4. Washington, D.C.: U.S. Government Printing Office.

• Vraeiu, Robert A., and Phyllis M. Virgil (1986) The Impact of Investor-Owned Hospitals on Access to Health Care. Paper presented at the Hirsch Symposium, The George Washington University, and published in Warren Greenberg, editor; and Richard M. F. Southby, editor. (eds.) For-Profit Hospitals: Access, Quality, Teaching, Research. Columbus, Ohio: Battelle Press.

• Watt, J. Michael, Steven C. Renn, James S. Hahn, Robert A. Demon, and Carl J. Schramm (1986) The Effects of Ownership and Multihospital System Membership on Hospital Functional Strategies and Economic Performance. This volume.

• Winn, Sharon (1974) Analysis of Selected Characteristics of a Matched Sample of Nonprofit and Proprietary Nursing Homes in the State of Washington. Medical Care 12(March):221-228. [PubMed: 4206470]

Appendix To Chapter 5. Data on Hospital Services and Facilities

TABLE 5.A.1 Percent of Hospitals with Various Services and Facilities, by Type of Ownership and Selected Bed Size Categories, 1983

*

Too few cases reporting to determine percentages.

SOURCE: Compiled from data provided by Hospital Data Center, American Hospital Association, Chicago, Illinois, 1985.

1. Estimates of the number of people who are unable to pay for medical care vary. National sample surveys in 1977 and 1982 showed that approximately 9 percent of the population—almost 20 million people—had neither health insurance nor eligibility for public programs, and as many as 9 percent of the insured population reported having been without insurance at some time during the previous year (Farley, 1985a; Robert Wood Johnson Foundation, 1983). More recent estimates of the uninsured population range as high as 15 percent of the population or 35 million people (Katherine Swartz, The Urban Institute, personal communication, 1985; U.S. Bureau of the Census, 1985). In addition, there are millions of underinsured people, whose limited insurance puts them at substantial risk of having out-of-pocket expenses upwards of 10 percent of their total income. The best data on this topic, the government's 1977 National Medical Care Expenditure Survey, found that depending on the definition used, from 5 to 18 percent of the population under age 65 was underinsured (Farley, 1985b).

2. A 1956 Internal Revenue Code Ruling held that for tax-exemption purposes, not-for-profit hospitals had to accommodate patients who were unable to pay, to the extent of their financial abilities. By 1983 that requirement had been dropped, as had another requirement for the operation of a full-time emergency department open to all patients, without financial prerequisites. A 1983 Internal Revenue Code Ruling (83-157) held that a hospital could maintain its tax exempt status in the absence of an emergency room if a state planning authority had determined that an emergency department would duplicate existing services. In that event, however, the hospital must—as community, service—accept Medicaid and Medicare patients, reinvest surplus revenue into capital improvements or health services, maintain an open medical staff, and appoint a governing board representative of the composition of the area (Bernstein, 1984). Other legal obligations flow from state and local interventions to ensure that hospitals offer some care for poor people. For example, under Texas law, hospital districts are responsible for their "needy," and public district and county hospitals are responsible for the "indigent sick." Furthermore, the Texas Property Tax Code states that to be tax-exempt, hospitals must be organized to perform a charitable purpose, generally by providing medical care without regard to the ability, of the beneficiaries to pay. The only requirement that applies to for-profit as well as not-for-profit hospitals in Texas is an obligation to provide emergency care in life- or limb-threatening circumstances.

3. Uninsured is defined as self-pay, reduced pay, Hill-Burton, or no charge.

4. Although bad debt and charity are conceptually different, the way they are accounted for and reported by hospitals is influenced less by conceptual distinctions than by reimbursement rules, Hill-Burton obligations, and other factors. Thus, the two figures have been combined in the committee's analyses reported herein. Deductions from revenue for bad debt and charity care exaggerate the cost to institutions of providing uncompensated care, because such cost is measured in terms of charges for services rather than the marginal cost of providing the services.

5. For example, the Hospital Corporation of America (HCA) reports that uncompensated care amounts to 3 percent of revenues in areas of Kentucky, where public facilities exist, but is 4.8 percent of revenues in areas where the HCA hospital is a sole provider (Vraciu and Virgil, 1986).

6. Only 11 for-profit hospitals listed in the Federation of American Hospital's 1985 directory appear among the 365 hospitals on the Health Care Financing Administration's (HCFA's) list of sole community hospitals as of July 17, 1985. Two of these institutions were independent. It should be noted that the HCFA definition of sole community provider is partly designed to minimize the number of such institutions, and that people in a larger number of communities may view their own hospital as the only one that is reasonably convenient and available.

7. No data are available describing changes that occur when hospitals are acquired by not-for-profit multi-hospital systems.

8. Hoy and Gray (1986) found that 80 percent of the hospitals owned by the largest six investor-owned hospital companies were either acquired from previous for-profit owners or were newly constructed. Acquisition of public or not-for-profit hospitals has become more common in recent years.

9. A design for a systematic study was recently explored by Jack Hadley and Judith Feder of the Georgetown University Center for Health Policy Studies. Their approach was to use data from American Hospital Association surveys between 1977 and 1981 and to focus on hospitals' (1) deductions from gross revenues for charity care and bad debt and (2) revenue from privately insured patients. The study was to focus on trends of these measures before and after the entry into a community of an investor-owned hospital or an existing hospital becoming part of an investor-owned or not-for-profit multihospital system. Unfortunately, the data proved to be inadequate, and the study could not be done.

10. Case examples illuminate the dynamics of such interactions among the hospitals in a service area. One such case study is the Public Broadcasting System's "Crisis at General Hospital," which describes the forced reduction in uncompensated care at a public hospital—Tampa General Hospital in Florida, owing to the hospital's loss of revenue-producing patients to for-profit and not-for-profit hospitals.

The committee's case studies (Townsend, 1986) also show what the loss of substantial numbers of paying patients can mean to hospitals committed to providing care to the poor and uninsured of their communities. Two of the case studies were conducted in cities that had particularly difficult indigent care problems. In both cities religiously affiliated hospitals that made serious efforts to provide uncompensated care had been affected by the construction of investor-owned hospitals that provided very modest amounts of uncompensated care. In one case the investor-owned hospital had only moderate success in drawing patients away from the existing two hospitals. Although the stronger of the existing hospitals felt relatively little impact from the addition of a new hospital (and continued its provision of uncompensated care and construction and renovation projects while maintaining a healthy surplus), the weaker existing hospital developed a deficit and closed its obstetrical unit because of lack of paying maternity cases. Whether these changes were due to competition from the investor-owned hospital or to other factors was not clear. However, the key role of surplus revenues to cross-subsidize uncompensated care is well illustrated by the contrast between the two hospitals.

The second case study illustrated more clearly how a new investor-owned hospital, if successful at drawing paying patients, can affect the existing providers. In this case two well-established religious hospitals experienced substantial census declines after an investor-owned hospital opened. The hospitals suffered a period of financial and management stress while rebuilding the census—an effort helped by a growing local population. Continued cross-subsidization of uncompensated care was undoubtedly helped by the existing hospitals raising their prices substantially, bringing them more closely in line with the new investor-owned hospital. Such options will become less feasible if more price competition develops.

11. There are many reasons other than profitability why an institution might not offer obstetrical services. It hardly makes sense for all hospitals to offer obstetrical services, particularly with birth rates having declined. Furthermore, because a significant portion of investor-owned hospitals are new facilities constructed after the implementation of health planning and certificate-of-need programs, some investor-owned hospitals may not have been allowed to offer obstetrical services because of the availability of such services at other hospitals.

12. This comparison does not indicate the size of premature nurseries. Some hospitals may have very small premature nurseries for temporary care of babies prior to transfer.

13. The six chains are Hospital Corporation of America; Humana, Inc.; American Medical International, Inc.; National Medical Enterprises; Charter Medical Corporation; and Republic Health Corporation.

14. Lending support to the interpretation that it is the proprietary, rather than the investor-owned, for-profit hospitals that account for most for-profit hospital closures is the fact that while Sloan et al. (1986) found that 50 "for-profit" hospitals had closed during the period 1980-1982, Hoy and Gray (1986) found that only two hospitals owned by the six largest investor-owned hospital chains had closed during that period.

15. Nursing homes are generally paid a per-diem rate, regardless of the cost of caring for individual patients. A few states have attempted to improve access for heavycare patients by paying a higher rate for patients needing more intensive care, but in general there is a disincentive to admit heavy-care patients. In Illinois, where "points" are awarded for a patient's disabilities, there are fears that nursing homes have "gamed" the system by making patients appear to be more sick than they really are.

16. Some states have acted to discourage such schemes. For example, Minnesota prohibits charging higher rates to private-pay patients. Massachusetts prohibits the refusal of a Medicaid patient when a bed is available. Connecticut requires admission on a first-come, first-served basis.

17. For example, a study in Washington State found no difference by ownership type in the use of nursing homes by Medicaid patients (Winn, 1974), while a study in the Cleveland metropolitan area showed that for-profit homes served a higher proportion of Medicaid patients than did not-for-profit homes (Brooks and Hoffman, 1978).

18. Health planning agencies, established under the National Health Planning and Resources Development Act of 1974, have in the past done important work in assessing the adequacy of access to care at the state and local levels. Although these agencies have undergone severe funding cuts, all states have a planning agency today and 141 local agencies still exist (Terry Shannon, Director of Field Services and Private Sector Programs, American Health Planning Association, personal communication, 1985). In principle, planning agencies at the state and local levels are well situated to provide important forums of discussion and information collection and dissemination regarding issues of access to care.

19. Table 5.2 shows that contributions amount to 0.4 percent of the revenues of not-for-profit hospitals and 1.0 percent of the revenues of public hospitals. However, even these numbers can be misleading. A recent American Hospital Association (unpublished data, 1983) survey of sources of working capital found that only 42 percent of responding hospitals reportedly received philanthropic support and that the median amount of support for these hospitals was $43,700. (The mean amount was more than $700,000. The wide gap between mean and median is due to the influence of very large philanthropic contributions received by a relatively few institutions. Thus, the median provides a better indicator of what the typical institution might have received.) An earlier American Hospital Association (1979) survey showed that 71 percent of the respondents received charitable donations—the mean amount was just over $200,000. (Unfortunately, no median figure was available.) Neither of the above surveys shows the amount of philanthropy available for general operating purposes, which is where money to subsidize uncompensated care would presumably come from. However, a 1984 survey conducted by the National Association for Hospital Development suggests that only a small amount of the charitable contributions received by hospitals is available for general operating purposes (AAFRC, 1985). The Association's 1,500 individual members at 1,200 hospitals (presumably the bulk of institutions that have an organized fund-raising apparatus) reported that they had raised just over $1 billion in 1984, and that 12.7 percent of this money, a mean of $108,000 per institution, was for general operating purposes. (Funds for construction and renovation comprise 25 percent of the total; spending for equipment, 17 percent; and research and education, 9 percent.) Given that the average hospital's total net revenues in 1983 were in excess of $21 million (American Hospital Association, 1984)—and the hospitals that had formal fund-raising activities were presumably larger than the average—charitable contributions for general operating purposes averaged less than 1 percent of hospital operating revenues. Given the fact that relatively few hospitals benefit from very large charitable contributions, charity for general operating purposes would appear to be less than 1 percent of revenues at most institutions, a conclusion that is supported less inferentially by Table 5.2.

20. There was substantial variation among the four companies—from 1.1 percent for National Medical Enterprises to 4.4 percent for Humana. For-profit hospitals pay property taxes in addition to income tax. It is not known how much property tax is paid, although an industry source estimates it at roughly 20 percent of income taxes (Samuel Mitchell, Federation of American Hospitals, personal communication, 1985). This would inflate the tax burden above the figures shown in Table 3.7 in Chapter 3.

21. Actions that can be taken to enhance access to care for disadvantaged people (some of which are being considered or are being acted on in various states) include creating a funding pool with which hospital care can be financed. A pool can be created by taxing all hospitals, or only those hospitals that fail to provide a specified amount of uncompensated care. Florida has established a pool through a contribution from state general revenues and an assessment on the net operating revenues of hospitals. Tying a hospital's assessment to the amount of uncompensated care provided reduces the incentive to "dump" patients, as do proposals requiring hospitals to devote a specified percentage of revenues to indigent care.

Some states (including Maryland, New Jersey, and Massachusetts), with all-payer rate setting, include charity or bad debt allowances in payments to hospitals. Other states create funding pools through a tax on health insurance premiums or by earmarking certain portions of general sales or other taxes. States have the option of using pooled money to finance care for all "medically needy" people or of targeting funds for especially vulnerable populations. Florida may use its funds to provide care for specific groups that are not covered by Medicaid.

Groups can also be targeted through direct governmental payments to certain teaching hospitals or governmental hospitals (Colorado, Virginia, North Carolina) or through an expansion of state Medicaid programs. In 1984, South Carolina extended Medicaid coverage to low-income pregnant women, regardless of marital status, and to children under 18, regardless of the marital status of their parents. Some states developed insurance programs to cover "catastrophic" episodes of sickness, unemployed people, and those unable to obtain conventional insurance. Finally, to spread the burden of uncompensated care more evenly among providers, licensure or certificates of need could be contingent on providing certain amounts of uncompensated care. Each approach differs in terms of who pays (e. g., hospitals, insurers, taxpayers), who benefits (e.g., people with specific medical needs such as pregnancy care, groups defined by income, groups defined by gaps in insurance), and the incentive to providers (e.g., to provide care for targeted groups, to provide uncompensated care generally, or reduce "dumping").

In seeking to reduce the number of people who are unable to find care, an option that might be considered is the use of tax credits or waivers to encourage for-profit health care institutions to provide more care to people who are unable to pay or to provide services that may be needed in a community, but that cannot be provided profitably. The use of the federal tax law to encourage private corporations to act in a way that advances some important public policy goal is hardly unprecedented, and local taxes are frequently waived or adjusted to encourage corporations to locate in a given area. However, little attention has been given to the possible use of taxing power to encourage institutions to meet public purposes.