Health Affairs Specialty Hospitals : A Problem Or A Symptom ?

The Medicare Prescription Drug, Improvement, and Modernization Act (MMA) of 2003 imposed a moratorium on new physician-owned specialty hospitals, which expired in June 2005. Recent administrative decisions have effectively extended the moratorium as it applies to Medicare payment, and the debate continues over whether specialty hospitals are good, bad, or indifferent. This paper reviews the findings in two congressionally mandated reports on specialty hospitals, the services they provide, and their impact on their patients and communities. These findings are used to highlight the major policy issues to be addressed—issues that go beyond whether specialty hospitals should be allowed. [Health Affairs 25, no. 1 (2006): 95–105] I n the medicare pre scr i pt ion drug, Improvement, and Modernization Act (MMA) of 2003, Congress imposed an eighteen-month moratorium on physician referrals to cardiac, orthopedic, or surgical specialty hospitals in which the physician has an ownership or investment interest. At the same time, Congress required that the Medicare Payment Advisory Commission (MedPAC) and the secretary of health and human services (HHS) study and report on various aspects of the performance of specialty hospitals, referral patterns for the types of patients they treat, their interactions with and impacts on local community hospitals, and their contributions to their communities. The MMA moratorium expired 8 June 2005, but the Centers for Medicare and Medicaid Services (CMS) effectively extended it through an administrative decision to “scrutinize whether specialty hospitals meet the definition of a hospital” and “carefully review our criteria for approving and starting to pay new specialty hospitals,” a process that will extend into 2006. The controversy over specialty hospitals doubtless will continue long after that. This paper reviews the context for this debate—including why specialty hospitals have been the object of such concern—and the findings and recommenP r o b l e m O r S y m p t o m ? H E A L T H A F F A I R S ~ V o l u m e 2 5 , N u m b e r 1 9 5 DOI 10.1377/hlthaff.25.1.95 ©2006 Project HOPE–The People-to-People Health Foundation, Inc. Stuart Guterman ([email protected]) is senior program director of the Program on Medicare’s Future at the Commonwealth Fund, based in Washington, D.C. He previously was director of the Office of Research, Development, and Information at the Centers for Medicare and Medicaid Services and deputy director of the Medicare Payment Advisory Commission. on Jne 2, 2017 by H W T am H ealth A fairs by http://conealthaffairs.org/ D ow nladed fom dations presented in the reports by MedPAC and HHS. The paper also attempts to lay out the issues and evidence that must be considered in coming to an appropriate decision on the future treatment of specialty hospitals. Background On Specialty Hospitals Specialty hospitals are nothing new. Children’s hospitals have long been recognized as leaders in the development and delivery of health care to young patients and as assets to their communities and the nation; rehabilitation hospitals also are looked to as sources of specialized care for patients who are temporarily or permanently disabled; psychiatric hospitals address the needs of patients with mental conditions or alcoholor drug-related problems that require short-term acute care; eye and ear hospitals are viewed as centers for state-of-the art treatment for eye and ear conditions; and a small group of cancer hospitals are generally regarded as being at the forefront of cancer treatment. All of these groups of hospitals treat patients who otherwise generally would be treated in community hospitals, and, in fact, many community hospitals have units within them that are set up to provide treatment to these same groups of patients. Even among community hospitals, specialization in providing services that they can provide best and most efficiently has been viewed as a good thing. In its 1982 report to Congress laying out its proposal for an inpatient hospital prospective payment system (PPS), HHS listed as one of the system’s advantages that it “will result in improved quality of care as hospitals begin to specialize in what they do best.” The incentive to specialize under a fixed per case payment rate not only was recognized, but also was viewed as a potentially beneficial outcome. Concerns About Specialty Hospitals So why are there strong objections in some quarters to the development of cardiac, orthopedic, and surgical hospitals? There seem to be four major concerns, which are interrelated to varying degrees, but they begin with the issue of ownership by physicians. Clinical decisions made by physician-owners might be distorted by financial incentives. Physicians’ ownership of specialty hospitals raises the concern that physician-owners will be torn between considerations of clinical appropriateness and financial benefit when deciding which treatment options and settings to recommend. Charles N. “Chip” Kahn III, president of the Federation of American Hospitals, asserts that “these facilities provide strong financial motivation for physicianowners to keep profitable admissions coming because the facilities offer [them] exclusive and extremely lucrative financial returns unavailable to anyone else.” The potential for conflicts of interest in health care have long been recognized. In the Omnibus Budget Reconciliation Act (OBRA) of 1989, Congress prohibited physicians from referring patients to entities for clinical laboratory services if the physician or a member of the physician’s immediate family had a financial relation9 6 J a n u a r y / F e b r u a r y 2 0 0 6 S p e c i a l t y H o s p i t a l s on Jne 2, 2017 by H W T am H ealth A fairs by http://conealthaffairs.org/ D ow nladed fom ship with the entity. Subsequent legislation and regulations expanded the range of services covered under the provision (known as the Stark prohibitions) and clarified the circumstances under which it would be applied. Two important circumstances under which the Stark provisions do not apply are the “whole hospital” exception and the case of ambulatory surgery centers (ASCs). The “whole hospital” exception provides that physicians may refer patients to a facility in which they have an ownership interest if their ownership interest is in the whole hospital, rather than a smaller entity. The rationale for this exception is that an individual physician is likely to own such a small share of an entire hospital that the resulting financial incentive would be unlikely to cause a conflict of interest. Before MMA’s enactment, specialty hospitals were considered to be covered under this exemption. Those who oppose specialty hospitals, however, assert that they, in providing a limited range of services, are not whole hospitals and that physician ownership of them should not be allowed. ASCs also are exempt from the Stark provisions, although in many ways they appear to be subject to the same types of adverse financial incentives that apply to specialty hospitals. A study by MedPAC found that Medicare ASC payment rates are higher than the corresponding rates for hospital outpatient departments for eight of the ten procedure codes with the highest share of Medicare payments to ASCs, and that patients treated in ASCs are less medically complex than those treated in outpatient departments. Although supporters of this exemption argue that ASCs should not be considered comparable to specialty hospitals, any reconsideration of specialty hospitals’ status might well include ASCs, if only to clarify the notions of consistency of payment systems and potential conflicts of interest. Specialty hospitals treat less complex, more profitable cases. Because the physicians who own specialty hospitals can, to some extent, control the flow of patients who are admitted, competitor hospitals accuse them of siphoning off the least complex and most profitable cases. The relationship between complexity and profitability is a health care financing issue: If payment for each hospital discharge corresponds to the anticipated costliness of treating that type of case, there will be no financial penalty associated with treating more-complex and -costly cases and no incentive to focus on simpler, less costly cases. The lack of such correspondence between payment and anticipated costliness raises the prospect of adverse incentives: To the extent that certain hospitals can specialize in easier and less costly— and therefore more profitable—cases, they can benefit from a type of specialization that is entirely different from the focus on service efficiency and effectiveness that was envisioned when the Medicare PPS was developed. It must be noted, however, that among the almost 5,000 U.S. community hospitals, the mix of cases across clinical categories—and even the mix of cases within those categories—varies considerably, as does the way they are treated. It should not be presumed that community hospitals are a homogeneous group; therefore, it is difficult to compare specialty hospitals with community hospitals as a whole, in P r o b l e m O r S y m p t o m ? H E A L T H A F F A I R S ~ V o l u m e 2 5 , N u m b e r 1 9 7 on Jne 2, 2017 by H W T am H ealth A fairs by http://conealthaffairs.org/ D ow nladed fom either their inclination or their ability to respond to the potentially adverse incentives that might be presented by the payment systems they face. Specialty hospitals avoid patients who are uninsured or underinsured. Regardless of the ability of Medicare or private payers to align their payments with the anticipated costs of each type of case, there is a strong incentive for hospitals to avoid patients who cannot pay at all, or those, such as Medicaid patients in many states, for whom the payment rate is low relative to that of other patients. In the current health care financing system, uninsured or underinsured patients are not accompanied by an explicit and reliable promise of payment and therefore present the prospect of financial loss to providers who treat them. To the extent that hospitals can choose their locations, their admitting physicians, and other aspects that can affect the payer mix for patients they treat, there is a strong incentive to do so. Again, however, it must be noted that the differential inclination or ability to respond to these potentially adverse incentives is not merely a matter of the distinction between specialty hospitals and community hospitals as a group; there are dramatic differences among community hospitals themselves in the extent to which they are affected by uncompensated care or Medicaid losses. An early study by the Prospective Payment Assessment Commission (ProPAC), a predecessor body to MedPAC, found that 25 percent of community hospitals incurred losses on uncompensated care (net of offsetting payments) equal to 9.1 percent of their total hospital costs in 1991, while another 25 percent sustained uncompensated care losses equal to only 1.1 percent of their total costs. Similarly, both the treatment of Medicaid patients and the financial consequences were found to be distributed extremely unevenly across community hospitals. Patient selection by specialty hospitals hurts community hospitals. Community hospitals argue that the strong financial incentives to select more-profitable cases and patients, combined with specialty hospitals’ greater inclination and ability to act on those incentives, makes community hospitals financially vulnerable. According to this argument, community hospitals had counted on the profitable cases that are siphoned off by specialty hospitals to cross-subsidize community hospitals’ treatment of less profitable cases and patients, as well as other community services they provide. Cross-subsidies have always been a key part of the health care financing system, with hospitals using profitable services and patients covered by profitable payers to offset their losses from less profitable services and patients. But the patterns of cross-subsidies vary greatly across facilities, with a relatively small number of community hospitals bearing a disproportionate share of that burden. Whether specialty hospitals exacerbate that burden is a matter to be studied. Congressionally Mandated Reports As mentioned above, Congress, faced with pressure from both voluntary and proprietary hospital groups to address the issue of specialty hospitals, imposed an 9 8 J a n u a r y / F e b r u a r y 2 0 0 6 S p e c i a l t y H o s p i t a l s on Jne 2, 2017 by H W T am H ealth A fairs by http://conealthaffairs.org/ D ow nladed fom eighteen-month moratorium on new cardiac, orthopedic, and surgical facilities and called for two reports to help address the issue when the moratorium was to end 8 June 2005. Between the two reports, Congress was looking to determine the answers to several questions that have been central to the debate over the merits and impact of specialty hospitals: whether specialty hospitals are more efficient than community hospitals, whether they deliver care of comparable quality, whether they engage in selection of less complex and more profitable cases, whether they undermine the financial viability of community hospitals, and whether they contribute to the community’s benefit. The MedPAC report. MedPAC analyzed data from Medicare hospital cost reports and inpatient claims for 2002, the latest year for which data were available to them, supplemented by site visits. MedPAC’s analysis was based on forty-eight physician-owned specialty hospitals (twelve heart hospitals, twenty-five orthopedic hospitals, and eleven surgical hospitals) that met specified criteria. Study findings. Specialty hospitals—particularly orthopedic and surgical hospitals—tend to be small, with an average of sixteen beds at orthopedic hospitals and fourteen at surgical hospitals. Cardiac hospitals tend to be somewhat larger, with an average of fifty-two beds. MedPAC found that, in contrast with community hospitals, many specialty hospitals do not have emergency departments; this was particularly true of the orthopedic and surgical hospitals it analyzed. Moreover, specialty hospitals are not evenly distributed across the country: Almost 60 percent are located in four states (South Dakota, Kansas, Oklahoma, and Texas), and almost all are located in states without certificate-of-need (CON) programs. MedPAC found that physicians had several motivations for investing in specialty hospitals. An important consideration was control over hospital operations, which allowed them to avoid disruptions to their ability to schedule time in the operating room, increase the operating room’s efficiency, and work more effectively with the staff. Income enhancement was another motivation; in addition to more productive use of hospital resources, the opportunity to retain a share of facility profits was an attractive prospect. MedPAC found that annual returns frequently exceeded 20 percent of physicians’ initial investment and that the average all-payer margin at the hospitals in their study was 13 percent, compared with 3–6 percent for community hospitals in the same markets. The MedPAC analysis, however, could not attribute this difference in financial performance to greater efficiency at specialty hospitals. When MedPAC compared case-mix-adjusted costs per discharge among specialty hospitals and other comparable groups of hospitals in the same markets, no statistically significant differences among the pairs of groups were found. MedPAC also found that physician-owned specialty hospitals tended to treat fewer Medicaid patients than community hospitals did in the same market, with cardiac hospitals treating primarily Medicare patients and orthopedic and surgical hospitals treating primarily privately insured patients. P r o b l e m O r S y m p t o m ? H E A L T H A F F A I R S ~ V o l u m e 2 5 , N u m b e r 1 9 9 on Jne 2, 2017 by H W T am H ealth A fairs by http://conealthaffairs.org/ D ow nladed fom Specialty hospitals obtained most of their patients by capturing market share from community hospitals, but this did not appear to have had a substantial effect on the financial performance of the community hospitals. MedPAC pointed out, however, that the number of specialty hospitals in its analysis was small and they were relatively new, so their financial impact on community hospitals might well increase over time. Finally, MedPAC found that specialty hospitals tended to treat more-profitable patients. There are two ways for this to occur: If the Medicare payment rates, which correspond to clinical categories (diagnosis-related groups, or DRGs), do not adequately reflect the relative costliness of cases across those categories, some DRGs will be more profitable and some less so; also, if groups of cases within a given DRG predictably differ in complexity and costliness, the less complex and costly cases will be more profitable (because the hospital receives the same payment for all cases in the DRG). Notably, cardiac hospitals appeared to treat cases in more-profitable DRGs relative to community hospitals but did not appear to treat cases that were significantly less complex within DRGs. Surgical hospitals, by contrast, appeared to have treated about an average mix of cases across DRGs, but the cases they treated were less complex than other cases in the same DRGs. Orthopedic hospitals actually treated patients in less profitable DRGs than did the community hospitals in the same markets, but their patients tended to be much less complex than other patients within those DRGs. Recommendations. To improve payment accuracy both across and within DRGs, MedPAC called for refining the current DRGs, to capture more completely the differences in complexity among groups of cases within the current DRGs, and revising the way the DRG relative weights are calculated, to reflect more appropriately the relative costliness of cases across DRGs. These recommendations are not new; MedPAC and its predecessor, ProPAC, have long been recommending both the refinement of DRGs and the revision of the methodology for calculating the relative weights. In fact, MedPAC’s recommendations in this report mirror almost exactly ProPAC’s recommendations made in 1995. This similarity indicates how long the payment system has needed revising—and how difficult it has been to get those needed changes implemented. Recognition of this difficulty is indicated by MedPAC’s additional recommendation that the DRG changes be implemented over a transitional period. MedPAC also made two recommendations that address the issue of specialty hospitals more specifically: (1) that the moratorium on specialty hospitals be extended until 1 January 2007, to allow Congress time to consider and develop legislation in response to its recommendations and to give the CMS time to respond; and (2) that Congress give HHS authority to allow gain-sharing arrangements between physicians and hospitals and to regulate those arrangements to protect the quality of care and minimize the financial incentives that could affect physicians’ referrals. This authority could be used to provide alternatives to the financial in1 0 0 J a n u a r y / F e b r u a r y 2 0 0 6 S p e c i a l t y H o s p i t a l s on Jne 2, 2017 by H W T am H ealth A fairs by http://conealthaffairs.org/ D ow nladed fom centives that have led to the increase in physician-owned specialty hospitals. The HHS report. The CMS analyzed referral patterns among specialty hospitals and their competitors—community hospitals and academic medical centers located within twenty miles of a specialty hospital—using Medicare claims data for 2003. The agency also assessed the quality of care at the study hospitals and their competitors using inpatient hospital quality indicators developed by the Agency for Healthcare Research and Quality (AHRQ). Data obtained from Internal Revenue Service (IRS) submissions and financial reports, as well as from the hospitals themselves, were used to estimate tax liability and uncompensated care. The CMS supplemented its analyses of administrative data with site visits to six market areas—Dayton, Ohio; Fresno, California; Rapid City, South Dakota; Hot Springs, Arkansas; Oklahoma City, Oklahoma; and Tucson, Arizona—where eleven (four cardiac, five orthopedic, and two surgical) of the sixty-seven physician-owned specialty hospitals that were operational in 2003 were located. Hospital executives, clinicians, managers, physician-owners, other physicians, emergency department staff, and finance staff were interviewed at each specialty hospital; executives at several competitor hospitals in each market area also were interviewed. In addition, patient focus groups comprising beneficiaries treated in specialty and competitor hospitals were used to assess patient satisfaction. Study findings. The site visits indicate that cardiac hospitals differ greatly from orthopedic and surgical hospitals and that discussions of specialty hospitals should reflect this dichotomy. Cardiac hospitals resemble full-service hospitals because of their size (fifty to eighty beds, with an average daily census, or ADC, of forty patients), the presence of emergency departments, and their community outreach programs. All of the cardiac hospitals in the study (sixteen had been operational for more than a year by 2003) had been built exclusively for cardiac care. They treated about 38,000 Medicare cases in 2003, about four times the number of Medicare discharges at surgical and orthopedic hospitals combined. Medicare accounted for 67 percent of all inpatient days at cardiac hospitals. In aggregate, physicians had a 34 percent ownership share in the cardiac hospitals in the study, although the average ownership share per physician was only 1 percent. Based on information gathered from all sixteen cardiac hospitals nationwide, the aggregate physician-ownership share was 49 percent, with a nationwide organization such as MedCath or a local nonprofit hospital typically owning the balance. Orthopedic and surgical hospitals were found to resemble ASCs more closely than they resembled hospitals, focusing on outpatient services (their aggregate ADC is only about five). Physicians tended to own the predominant interest, averaging 80 percent in aggregate for the orthopedic and surgical hospitals in the study, although the average ownership share per physician was only 2 percent. The remaining percentage interest typically was held by a nonprofit hospital or national corporation. Medicare accounted for about 36 percent of the inpatient days in orthopedic and surgical facilities. P r o b l e m O r S y m p t o m ? H E A L T H A F F A I R S ~ V o l u m e 2 5 , N u m b e r 1 1 0 1 on Jne 2, 2017 by H W T am H ealth A fairs by http://conealthaffairs.org/ D ow nladed fom Based on the ownership data obtained from the eleven hospitals in the site-visit markets, the study found that Medicare referrals to physician-owned specialty hospitals came primarily from the physician-owners (48–98 percent of all Medicare admissions). However, physician-owners did not necessarily admit their patients exclusively to their own hospitals. At the three cardiac specialty hospitals visited in the CMS study, the percentage of their own patients admitted by physician-owners to the specialty hospitals were 50 percent, 65 percent, and 75 percent, respectively. Physician-owners at the orthopedic and surgical hospitals visited in the study tended to admit most of their patients to competitor hospitals, presumably because of the limited inpatient capacity at their own hospitals. The Medicare cardiac patients treated in community hospitals were more severely ill than those treated in cardiac specialty hospitals in most of the study sites. However, there was much variation in the average severity level among patients across both cardiac hospitals and their competitors. This generally was true for patients admitted both by physician-owners and by other physicians, indicating no difference in referral patterns by ownership. Although the number of cases was too small to draw definitive conclusions for orthopedic and surgical patients, the same relationship appears to have held, but the difference in the proportion of severely ill patients was greater. Based on administrative data for all specialty hospitals nationwide in 2003, the proportion of patients transferred from cardiac hospitals to community hospitals was about the same as the proportion transferred between community hospitals. Moreover, the proportion of patients transferred from cardiac hospitals to community hospitals who were severely ill was similar to that of patients in the same DRGs who were transferred between community hospitals. These results provide no indication that cardiac hospitals transfer a greater number of severely ill patients to their competitors. Because of the small number of cases, no conclusions could be drawn about the severity levels of patients transferred between orthopedic or surgical hospitals and community hospitals. Based on an analysis of claims data using the AHRQ indicators and methodology, cardiac hospitals appear to have delivered care that was as good as or better than care at their competitor hospitals. Because of the small number of discharges, a statistically valid assessment could not be made for orthopedic or surgical hospitals. Patient satisfaction was very high for specialty hospitals, as Medicare beneficiaries reported enjoying large private rooms, quiet surroundings, adjacent sleeping rooms for their family members if needed, easy parking, and good food. Some of the hospitals in the site-visit markets provided financial information that allowed the computation of their taxes paid and uncompensated care as a proportion of total operating revenues. Among those facilities, the share of uncompensated care costs was much lower among specialty hospitals than among competitor hospitals. On the other hand, competitor hospitals received offsetting disproportionate-share hospital (DSH) payments from Medicare and Medicaid, 1 0 2 J a n u a r y / F e b r u a r y 2 0 0 6 S p e c i a l t y H o s p i t a l s on Jne 2, 2017 by H W T am H ealth A fairs by http://conealthaffairs.org/ D ow nladed fom while specialty hospitals paid real estate and property taxes as well as income and sales taxes. As a result, the total proportion of net revenue that specialty hospitals devoted to uncompensated care and taxes combined exceeded the proportion of net revenues that community hospitals devoted to uncompensated care. HHS recommendations. The HHS report to Congress contained no recommendations, but in subsequent testimony, Mark McClellan, CMS administrator, outlined four responses to the information contained in both the HHS and MedPAC reports. McClellan expressed agreement with the MedPAC recommendations that the accuracy of Medicare payment rates needs improvement. The CMS will be analyzing the potential impact of various options, including those recommended by MedPAC, and will incorporate the results in proposed changes for fiscal year 2007. McClellan also indicated that the CMS is working to revise the payment rates for ASCs, which might reduce physicians’ incentives to participate in the ownership of specialty hospitals. MMA requires that the new ASC payment system be implemented no later than 2008. The CMS also is examining whether entities that currently have or are applying for a Medicare provider agreement meet the definition of a hospital for purposes of Medicare payment. The key requirement in this context is that a hospital must be engaged primarily in furnishing inpatient services. The CMS study indicated that many specialty hospitals, particularly orthopedic and surgical hospitals, concentrate primarily on outpatient care. In a related action, the CMS is planning to consider how to apply the anti-dumping requirements of the Emergency Medical Treatment and Active Labor Act (EMTALA) of 1986 to specialty hospitals and other facilities. The process was expected to be completed early in January 2006.