Rate Regulation as a Policy Tool: Lessons From New York State

Abstract

For over a decade, New York State has used hospital rate regulation (the New York Prospective Hospital Reimbursement Methodology [NYPHRM]) as a policy tool to achieve three objectives: containing costs, supporting financially stressed hospitals, and financing access to care for the uninsured. This case study of NYPHRM suggests that the regulatory approach, if pursued with vigor, can achieve any one of these goals. On the other hand, the New York experience also shows that these are competing goals, and that achieving all of them over a period of time can prove to be difficult.

Introduction

In their search for ways to both contain health care costs and address the health care needs of the poor, States have taken several approaches. Some, such as Washington and Minnesota, are looking to comprehensive reform of the overall system; others, such as Tennessee and Oregon, are hoping that cost savings under managed care will enable the Medicaid program to cover more people at lower cost; a third group, including California and Florida, is looking to some form of managed competition as the primary vehicle for controlling costs. This article, a case study of New York's experience with hospital ratesetting, examines a fourth route—use of the State's regulatory powers to both limit costs and expand access. This study is based on a review of published analyses and formal evaluations, plus extensive onsite interviews with those who have been most active throughout the history of the NYPHRM.

Since 1983, NYPHRM has evolved, matured, and evolved again many times over, and is now in its fifth iteration. As shown in Table 1, the NYPHRM experience includes 3 years (1983-85) of all-payer inpatient hospital ratesetting and several almost-all-payer methodologies that continue to the present. New York therefore provides one of the most mature models of a regulatory approach to cost containment. Perhaps more important, it also sheds light on the use of ratesetting as a policy tool for achieving a variety of health care objectives, including the generation of revenue to finance care for the poor.

Table 1. Historical Summary of the New York Prospective Hospital Reimbursement Methodology (NYPHRM) Models: 1983-95.

Model Iteration	Description
NYPHRM I (1983-85)	An all-payer system operating under a Medicare waiver, with hospitals paid on a per diem basis and all payers contributing to bad debt and charity care pools.
NYPHRM II (1986-87)	Medicare moved to case-based reimbursement, while the rest of the payers remained under a regulated per diem system. A surcharge on per diem payments collected money for the pools.
NYPHRM III (1988-90)	Other payers moved to case-based reimbursement, but continued to be separate from the Medicare prospective payment system.
NYPHRM IV (1991-93)	Continued the case-based system, but included adjustments to reflect new costs, and also introduced some broader insurance and delivery reforms.
NYPHRM V (1994-95)	Continued the case-based system and expanded the other broader reforms.

SOURCE: Fraser, I., Health Systems Research, Inc., 1995.

No State today would be likely to replicate any of the NYPHRM approaches exactly. The health care environment has changed substantially in the past decade, so even New York probably would not end up with the same rate regulation methodology if it were starting today. In fact, even as the fifth phase of NYPHRM begins, key policymakers from around the State are beginning to consider its successor, and to debate how (and if) NYPHRM can serve as the base for broader health care reform, and how (and if) elements of the regulatory and competitive approaches can be combined. Nevertheless, the New York experience provides some very useful lessons for policymakers considering possible use or incorporation of a regulatory approach in their health reform plans. Probably the most important lesson is that the regulatory approach, if pursued with vigor, can achieve a variety of goals—including support for distressed hospitals, increased access to care, and cost containment. On the other hand, as the New York case also shows, these are competing goals, and achieving all of them over a period of time can prove difficult.

This case study is intended to answer the following six key questions about New York's regulatory approach, with a particular focus on what policymakers in other States might need to design, adopt, adapt, and implement such a program: (1) How did the approach come to be considered? (2) What was it designed to do? (3) What was required for it to be adopted and implemented? (4) How and why did it change over time? (5) How well did it work? (6) How does it fit with the new reform environment? The article concludes with a section summarizing lessons for other States.

Background

Two aspects of the New York case are critical to an understanding of how and why the system emerged and subsequently evolved. First, the purpose of New York's regulatory activity, and in particular the purpose of the all-payer system introduced in 1983, was not simply to contain costs but also to broaden access and support distressed hospitals. While controlling costs has been an important goal of New York regulatory activity since 1983, it has not always been the most salient one. Second, NYPHRM did not emerge or continue as an isolated regulatory experience. It built on past systems of rate regulation and was very much influenced by New York State's activist approach to health policy—including a very stringent certificate-of-need (CON) process dating from the early 1960s—which has long gone beyond the issue of inpatient reimbursement.

The origins of NYPHRM go back to the late 1960s, when the State began to regulate not only Medicaid payments but Blue Cross payments as well. Blue Cross had at that time a large market share and a policy of community-rating all individual and small group policies, whereas commercial competitors were pricing policies based on expected use and beginning to specialize in the sale of policies to younger, healthier population groups. In a competitive market, commercial insurers would increasingly have attracted the good risks, and Blue Cross would have been left covering sicker, more expensive individuals.

One way to prevent this competitive disadvantage for Blue Cross would have been to create a level playing field, i.e., requiring all insurers to community-rate their individual and small group policies (a step the State in fact took many years later), but this approach would not have been politically feasible in the 1970s. Instead, New York chose an alternative way to achieve this same goal: giving Blue Cross a price break on hospital payments, so that Blue Cross could price its policies competitively despite the fact that many of its enrollees were sicker and potentially more costly (Padgug, 1991). Beginning in 1970, the State moved both Medicaid and Blue Cross to a per diem prospective payment system (PPS) and “coupled” their rates, meaning that the same methodology (prospective payment using cost data from 2 years before) would be used in calculating the rates for each of them. While the resulting rates were not identical, they were generally very close. Medicaid and Blue Cross rates were kept low, at or under costs, and commercial payers cross-subsidized both this arrangement and any recovery of costs for uncompensated care.

During the 1970s, however, this arrangement began to unravel, because the cross-subsidy needed was growing but the base of commercial payers was not. Part of the problem resulted from growth in the number of uninsured (and therefore the amount of uncompensated care). Equally important, however, was the growing cost of Medicaid. Medicaid costs in New York State have traditionally been very high compared with other States, because the State has one of the most generous Medicaid programs in the country, with high eligibility levels and coverage of almost all federally permitted services. (The Federal Government bears about one-half of this cost, and the rest is shared by State and local governments.) The escalation of costs in the late 1970s contributed to a series of severe budget crises both for the State and for New York City. To address this crisis, the State froze income eligibility levels and cut reimbursement rates for Medicaid. Because Blue Cross rates are coupled to those of Medicaid, reimbursement rates paid by Blue Cross fell as well.

Hospitals responded to the reimbursement decline in three ways: cutting costs, running deficits, and shifting costs to unregulated payers. The second and third strategies each had a destabilizing effect on the State's health care system. Deficit spending led some of the most prestigious and critical hospitals to the brink of financial ruin, with hospital operating losses reaching 3 percent of assets in 1977 and 1978 (Thorpe, 1987); cost shifting caused a rapid spiral in premiums for commercial insurers. By 1978, inpatient rates for commercial insurers were, on average, 25 percent higher than those of Blue Cross (Thorpe, 1987), and commercial insurers threatened to stop doing business in the State. While Blue Cross benefitted from this rate difference, this insurer faced some fiscal problems as well. Given the large and growing gap between Blue Cross reimbursements and the rates paid by commercial insurers, hospitals successfully argued in court that they should be permitted to bill patients for the balance remaining after payment by Blue Cross. In response, the State took a first step toward regulating commercial rates, capping the Blue Cross-commercial payment differential, with a view to reducing it over time. In search of a more lasting and comprehensive solution, the State also created the Council on Health Care Financing, a joint legislative body charged with developing a new system for financing hospital care.

Design and Adoption of the Approach

Major Goals and Design Issues

As previously suggested, the Council on Health Care Financing's charge was not simply to contain costs, but to design a hospital reimbursement system which would achieve several related (but also competing) objectives: curbing the growth in hospital costs, supporting distressed hospitals and spreading the cost of uncompensated care, and facilitating access to care for the poor.

Curbing Growth in Hospital Costs

Several players were particularly concerned about increasing health care costs: Employers were worried about the impact on premiums; commercial insurers were nervous because they were absorbing cost shifts; and both State and local governments were distressed about rising Medicaid costs. Because hospital care accounted for the bulk of these health care costs—particularly for employers, Blue Cross, and commercial insurers—discussion was focused on how to control inpatient reimbursements.

Faced with this scenario, there are two basic political choices a State might make on the reimbursement side: It can focus on controlling Medicaid reimbursements, letting providers shift costs to other payers in order to make up for any losses, or it can try to control overall costs by limiting provider reimbursements by all payers. Historically, most States have gone the first route, prompted by the need to balance their own budget. But by the time the Council on Health Care Financing began to meet it was clear that this approach would not work for New York, for financial, philosophical, and political reasons.

Financially, given the size of the New York State Medicaid program, and the coupling of Medicaid reimbursement rates with Blue Cross rates, holding down Medicaid reimbursements alone would dramatically expand costs for other payers, as had, in fact, already happened. Philosophically, New York objected to the Medicaid-only approach because of concern that major disparities in reimbursement between public and private payers would create a two-track system and therefore inhibit access for Medicaid patients. Finally, a two-track system would be less tenable politically in New York than in most States because the New York Medicaid program has a larger, more economically diverse, and therefore more powerful constituency than in most States, due in part to its higher eligibility levels (Fossett, 1993).

Supporting Distressed Hospitals and Spreading Uncompensated Care Costs

Hospitals in New York, as in all States, are major employers, and therefore unions, local governments, and citizen groups generally want to see them survive. But the desire to support distressed hospitals was broader and deeper in New York, in part because past CON activity had made individual hospitals more critical to their communities. New York had few excess beds in the late 1970s, so the financial failure of a hospital could create very significant access problems. While all economic groups would experience these problems, the loss of access could be particularly severe for the poor, as discussed later.

With the deterioration of hospital finances in the late 1970s, the State already had responded several times to prop up individual hospitals that were in financial jeopardy. But policymakers sought to replace these ad hoc bailouts with a more systematic solution to hospital financial problems. A related goal was to stabilize the insurance market by assuring the survival of Blue Cross and ending the need for hospitals to shift costs. The complicating factor from a design viewpoint, however, was that any policy involving a redistribution of funds to hospitals could potentially transfer money from one part of the State to another and therefore run afoul of regional rivalries in the State.

Facilitating Access to Care

Supporting distressed hospitals was also the means for achieving a third, closely related goal: Making sure that the poor and uninsured had access to hospital care. In New York, providing services to the needy is seen as an important part of the mission of the State, and health care for the poor is a central part of this package of services. For historical reasons (which include low Medicaid reimbursements to physicians as well as a shortage of physicians in indigent neighborhoods), hospitals in many areas serve as the main safety net for access by the poor to primary as well as inpatient and emergency care. The hospitals that were experiencing the greatest amount of financial distress tended to provide the greatest amount of care for the poor, and also tended to have the fewest commercial patients to whom costs could be shifted. The failure of these hospitals would obviously cause a major hole in the health care safety net.

Obviously, these three goals are interdependent: Support for distressed hospitals can facilitate access by the poor, and limitations on hospital cost growth can hold down insurance premiums, thereby also facilitating access by enabling more of the near-poor to obtain coverage. As the following section shows, the original design of NYPHRM was intended to address all three goals. At the margin, however, the goals can be competing. In particular, as Thorpe (1987) notes, use of the ratesetting process to generate revenues used to achieve “distributive” goals, such as financing uncompensated care, can conflict with the goal of limiting rates. For this reason, much of the subsequent history of NYPHRM involves policy swings among strategies designed to maximize one or another of these competing goals.

Summary of the Approach

NYPHRM I, approved in 1982 and enacted in 1983, was an all-payer system of State-regulated per diem rates, including eight important features:

• All payers were included—The previous regulation of Medicaid and Blue Cross rates was extended to commercial payers and (under a Federal waiver) to Medicare.

• Payments were prospectively determined, on a per diem basis—Under the new system, NYPHRM paid all hospitals on a prospectively determined per diem basis, with lower rates for Medicaid and Blue Cross. Although the subsequently enacted Medicare PPS provided for case-based reimbursement, New York's waiver permitted the State to continue using the per diem system for all payers, including Medicare.

• All payments were linked to the 1981 base year—Determination of each year's per diem rates for inpatient care was based on a complex formula which began with “allowable” 1981 costs trended forward to 1983 using an economic trend factor and divided by the number of patient days in 1981. A similar process was followed for 1984 and 1985. (Capital costs were paid separately.) While the definition of allowable changed over time, the base year did not. With each revision of NYPHRM, the key issues for all payers and providers concerned were which 1981 costs would be considered allowable and what assumptions would go into calculation of the trend factor.

• Payment rates for hospitals were linked in part to the costs of similar hospitals—In calculating payment rates, hospitals were first placed in peer groups (based, for example, on teaching status, location, service and case-mix variables, etc.). Hospitals with high costs relative to their peers saw a disallowance of some of their costs, and therefore a tighter reimbursement environment (Thorpe and Phelps, 1990). In other words, the amount of reimbursement was linked in part to the costs of similar hospitals, rather than simply the cost of an individual hospital. (Over time, the individual hospital cost component in the formula has been lessened, and greater weight given to the peer group cost component.) The intent was to bring costs at high-cost hospitals down to the average of others in the group, while at the same time limiting cost growth in the group as a whole.

• Payment rates differed among payers—Commercial insurers continued to pay hospitals higher rates than Medicaid and Blue Cross, but the amount of this differential was established by statute. The 25-percent differential existing prior to enactment of NYPHRM was reduced to 15 percent (Thorpe, 1987), in an effort to limit the cost-shift burden for commercial insurers.

• The cost of care to the poor and uninsured was financed through a system of pools—NYPHRM established a system of pools in which money collected from add-ons to inpatient rates was used to finance inpatient and outpatient hospital care for the poor and uninsured, and to ease hospital transition into the new system (Caligiuri, 1993; Thorpe, 1988; Fraser et al., 1990). Eight regional bad debt and charity care pools were created, financed by surcharges on third-party payers.¹

• Rochester's unique program was exempted from NYPHRM—The Rochester metropolitan area has a unique health care system which was exempted from NYPHRM under State authority to approve demonstrations. The Rochester area has strong regional planning, a single community-rated area-wide risk pool, a 70-percent market share by Blue Cross, extensive interhospital coordination and planning, stringent CON procedures, and high managed-care penetration. Most important, from 1980 to 1987, Rochester operated under a community-wide spending cap. During these years, the major payers (Blue Cross, Medicare, and Medicaid) provided hospitals a guaranteed annual budget. Subsequent evaluations (Leitman et al., 1993; U.S. General Accounting Office, 1993) have shown that this unique experiment has been very effective in containing costs.

• Regulatory authority rested with the Department of Health (DOH)—Rather than creating an independent commission to oversee the ratesetting process, New York opted to give primary responsibility to an existing governmental body, the DOH. In addition, two other preexisting entities contributed to this process. First, the State Hospital Review and Planning Council (SHRPC) formulates regulations related to the quality, efficiency, and economy of health care, which it then presents to the Commissioner of Health for approval. The SHRPC consists of 31 people appointed by the Governor with the consent of the Senate, representing consumers, providers, and others, and staffed by the DOH. Second, a four-member Independent Panel of Economists appointed by the Commissioner of Health is responsible for helping to formulate the year-to-year “trend factor.” Finally, the legislature itself has been very involved in determining the specifications for NYPHRM, spelling out details in the law which other States might handle through regulation.

Factors Critical to Adoption and Implementation

While the particulars of the original New York approach are not likely to provide a blueprint easily transferred to a different time and place, they can provide some important lessons for policymakers elsewhere about what it takes politically for an approach of this sort to be adopted and implemented. Those involved in the process cite the following eight factors as most critical.

Strong Leadership Over Time

The Council on Health Care Financing was small, consisting of nine members appointed by the Governor and six appointed by leaders of the State legislature. It was chaired by Senator Tarky Lombardi, with Assemblyman James R. Tallon serving as the vice-chair. In contrast to many study groups of this sort, the Council proved to be very energetic and focused and was able to achieve three rather uncommon goals: transcending institutional interests; crossing party and chamber lines; and focusing on big-picture policy questions. Much of the explanation for this outcome, according to individuals involved at the time, rested with the personalities of the individuals involved, and in particular on the high energy, extensive knowledge, and personal congeniality of Tallon and Lombardi. This same political leadership continued at the implementation stage.

Often the individuals most responsible for initiation of a new policy soon move on to the next challenge, and refining and implementing the policy then becomes the task of a new group (Sapolsky, Aisenberg, and Morone, 1987). The ensuing change in leadership frequently causes policies to fall apart at the implementation stage, particularly where issues are complex and technical and the dollars at stake are large. Participants interviewed cite the continued presence of Lombardi, Tallon, and others involved in the Council as a critical factor in the continuity of NYPHRM through a decade of challenges. A greater turnover in the legislature, particularly at the leadership level, would have made the continuation of NYPHRM much more difficult.

Support by Key Actors

NYPHRM represented a compromise involving several actors—the State health department, Blue Cross, commercial payers, local governments, and hospitals—and the support of all of these was critical to successful adoption of the approach. This support did not come easily. Hospitals, for example, were not unified on the issue: Suburban hospitals, which derived much of their income from private payers and had relatively low uncompensated care levels, were generally opposed, while inner city hospitals and others with high uncompensated care levels were supportive. Most important, the Hospital Association of New York State supported the concept because of the desire to support distressed hospitals and ensure access.

To maximize support from hospitals and local governments, the pooling mechanisms were carefully crafted. NYPHRM created separate regional charity care/bad debt pools and a separate public hospital pool in order to allay hospital (and legislator) concern that the program would shift funds from upstate to New York City or from private to public facilities (Thorpe, 1987).

Also important to success was obtaining a waiver from HCFA. To obtain this waiver, the State had to agree that the rate of increase in total hospital Medicare spending in New York State would not exceed the national average rate of increase for Medicare. If Medicare hospital spending in New York did exceed the national average growth rate, the State promised it would repay HCFA for any payments over the average (Thorpe, 1987).

Congruence With the State's Political Culture

Another factor important to legislative success was the fact that NYPHRM was congruent with New York's political culture and played to important values in the State, in particular:

• Perception of regulation as a legitimate State activity—In some States, the use of State powers to promote health or other social values is seen at best as a necessary evil. In New York, however, activity in these realms is seen as a legitimate and important function (Hackey, 1992). The State already had been very involved in regulating hospitals, so NYPHRM did not represent a radical departure from past policy, but rather an incremental change in the existing policy of rate regulation.

• Belief in institutional solutions, coupled with distrust of for-profit entities—NYPHRM also dovetailed with New York's institution-centered approach to political problem-solving. Particularly in health care, New York tends to build its policies around institutions, often big institutions. On the other hand, distrust of for-profit entities has also been an important part of the State's political tradition. The State has some of the most prestigious not-for-profit hospitals in the country, few for-profit hospitals, and one of the largest Blue Cross plans, so an approach which focused on maintaining the viability of hospitals and Blue Cross was politically well suited to the State. As discussed later, the emphasis on hospitals is changing in New York, as in other parts of the country, and later iterations of NYPHRM began to move funds to primary care. But particularly at the outset of NYPHRM, and even to this day, New York has had a greater institutional focus than most other States.

Technical Capacity and Administrative Expertise

A fourth advantage in New York was the development of a very sophisticated and experienced group of administrators in the DOH, with expertise not only in ratesetting but in other regulatory activities and broader health issues as well. Many of the key administrators in New York State's DOH have been involved in the process for over a decade (Hackey, 1993).

Data Capacity

Regulation also requires an ability to collect and analyze hospital data at the institutional and even departmental level. While most participants in the New York system do not feel they have an adequate amount of data—particularly on the outpatient side—they have far more information than most States, and it has taken a long time to develop this capacity.

Surviving or Avoiding Legal Challenges

An increasing problem for most State reform efforts is legal challenges under the Employee Retirement Income Security Act (ERISA) of 1974 (Public Law 93406), but so far New York's efforts have managed to survive the worst of these. The redistributive features of the New York system are particularly vulnerable to challenge. A recent court ruling in New Jersey (United Wire et al. v. Morristown Hospital et al.) called into question systems which permit different reimbursement levels by different payers or require payers to contribute to a pool for the uninsured. While much of this decision was reversed on appeal to the Third Circuit Court, the original decision in the meantime encouraged a proliferation of similar suits in New York, which is in the Second Circuit and therefore not bound by the decisions of the Third Circuit.

At the moment, use of the add-on in New York is somewhat protected because of a national tax law provision (162N of the Internal Revenue Code), sponsored by Senator Moynihan, which states that employers cannot deduct the costs of employee health insurance coverage unless they comply with New York ratesetting legislation. However, this provision is being challenged in the courts.

Cross-Subsidies From Medicare

During most of the period covered under NYPHRM, the majority of New York hospitals have had a favorable Medicare margin, in part because of favorable payment levels for teaching hospitals. This positive margin has made it easier for hospitals to withstand shortfalls from other payers. In short, the Medicare program provided a cross-subsidy to other payers.

Fit With Complementary Policies

No policy, however comprehensive, works in isolation. In New York, a long-standing activist approach in planning and quality regulation strengthened the impact of reimbursement incentives. Because of New York's strong and stringent CON program, for example, the State's hospitals have been operating at or near capacity, and therefore could not easily react to per diem rate reductions by increasing volume. (On the other hand, the limitation in long-term care beds has probably impeded inpatient cost-containment efforts, since the reduced availability of long-term care beds tends to prolong the average length of stay in hospitals.)

Although New York had many features helpful to implementation, several respondents suggested one important disadvantage: the lack of political insulation. In contrast to Maryland, where authority over ratesetting was lodged in an independent commission, decisions about the structure and details of successive iterations of NYPHRM were made by the DOH and the legislature. The lack of an independent commission, coupled with the periodic recalculation of rate structures, reportedly encouraged the development of a very political process, in which every political actor with any connection to health became embroiled, studying analyses of the supposed “winners and losers” with each potential change in the complex formula. Several interviewees noted that the process created an atmosphere of continuing tension between the health department and the hospital community. Over time, a politicalization of the ratesetting process took place which detracted from a sense of legitimacy concerning the rates. It is interesting to note, however, that the process did not cause most people to question the right of the State to set these rates.

Evolution and Adaptation

The system recommended by the Council on Health Care Financing has continued in some form until the present, but has changed considerably. The original legislation had a 3-year sunset, and each subsequent version has had a 2- or 3-year sunset as well. As shown in Table 1, NYPHRM has evolved with each iteration in response to changing needs and political dynamics both within the State and at the Federal level. Three themes emerge in tracing the history of this evolution. First, the goal of cost containment has gone in and out of first position as a policy goal. Second, the reimbursement scheme has increasingly been used as a policy tool for achieving other health care goals unrelated to the issue. Third, in recent years, discussions of NYPHRM have increasingly focused on the question of how, and whether, the NYPHRM system can be linked to broader reform initiatives.

Response to Changing Priorities

While the three original goals—containing costs, supporting distressed hospitals, and assuring access—have continued to drive the system, the ranking among these was not always the same. The need to control costs was a driving force behind NYPHRM I, but cost containment was clearly only a secondary concern with NYPHRM II (1986-87). The major change between NYPHRM I and NYPHRM II was the removal of Medicare from the all-payer system, and the main reason for doing so was to bring additional Federal dollars into the State. Shortly after enactment of NYPHRM I, the Federal Government began a case-based PPS for Medicare. New York operated under a waiver from HCFA during the NYPHRM I period, and therefore did not come under the system. When the waiver expired in 1985, however, it became clear that hospitals could expect to receive at least $200 million more under PPS than with a continued waiver (Vladeck, 1993). To capture this additional money, New York converted its system from an all-payer system to an almost-all-payer system. Under NYPHRM II, Medicaid, Blue Cross, and other private payers continued to reimburse hospitals on a prospective per diem basis, while Medicare followed the same prospective case-based approach used in other States.

NYPHRM III (1988-90) brought another major change in the reimbursement methodology, moving all payers to a case-based system.² This time, however, the change in methodology also brought with it a concerted State effort to constrain reimbursement increases. The period covered by NYPHRM III was a low point for hospitals financially, because of low reimbursement levels as well as two additional external events. First, the crack cocaine and acquired immunodeficiency syndrome (AIDS) epidemics, along with the movement to deinstitutionalize mental health services, caused increased demand on the inpatient system, mostly due to the need for mental health, neonatal, and trauma services by the uninsured and Medicaid patients. Second, Medicare began to tighten some of its reimbursement rules as well, so the Medicare surplus began to erode (Altman and Garfink, 1990). As a result, New York hospitals saw operating losses of more than one billion dollars in 1988 (Vladeck, 1993). No longer able to make up these losses through cost shifts, hospitals used up reserves and/or went further into debt.

A related design change in NYPHRM III was the introduction of negotiated rates within the regulatory framework, permitting HMOs to negotiate rates, even if these fell below the rates of Medicaid and Blue Cross. Because HMO penetration at that time was still relatively small, the immediate impact was not sizeable. As discussed later, however, this decision to overlay elements of competition on a regulatory framework had a longer term destabilizing effect on the elaborate system of cross-subsidies embodied in NYPHRM.

By the time the legislature began to craft the NYPHRM IV program in 1990, hospitals were able to make a very good case that they were in severe financial straits, and the goal of stabilizing the hospital environment for the moment achieved preeminence over the goal of cost containment. In designing the new system, therefore, the legislature added about $400 million a year in hospital payments, mainly by incorporating new adjustments for labor and other expenses into the 1981 base year (Vladeck, 1993).

Use as a Policy Tool

A second overall trend in NYPHRM has been its increasing use as a policy tool for achieving other, broader health care goals. These broader goals include the following:

• Insurance expansion—NYPHRM III introduced, and NYPHRM IV increased, the use of pool funds to purchase insurance rather than direct care. For example, under the Child Health Plus program funded through pool funds, families could buy basic primary and preventive care coverage for any child under 13 years of age who was ineligible for Medicaid, not already covered under a private plan, and below 185 percent of the Federal poverty level. Premiums were set on a sliding scale, with subsidies of up to $20 million drawn from the Statewide Bad Debt and Charity Care Pool (Caligiuri, 1993).

• Expanding primary health care—NYPHRM III diverted a small amount of inpatient money to primary care, drawing on pool monies to fund hospital-based programs to expand primary health care services. NYPHRM IV authorized the Commissioner to make grants to general hospitals to subsidize primary care services which enhance the delivery of primary health services to medically underserved areas and medically indigent persons; improve coordination of primary health care services and outreach to medically indigent patients; and foster coordinated linkages among providers, including strengthened relationships between hospital-based providers and community-based primary health care services providers (New York State Senate Research Service, 1990). This program was expanded further under NYPHRM V.

• Determining and meeting community needs—Beginning with NYPHRM IV, voluntary non-profit general hospitals are required to prepare a “community service plan”—a written analysis of the community they serve and their plan for serving it—and to file this report with the Commissioner and the appropriate health systems agency. As part of this report, hospitals need to review their mission statement, solicit community opinion on the hospital's performance and service priorities, and demonstrate the hospital's operational and financial commitment to meeting community health care needs identified by the health systems agency, providing charity care services, and improving access to health care services by the uninsured. Finally, hospitals are required to prepare and make available to the public a statement showing the financial resources of the hospital and related corporations and how these resources are allocated among various priorities, including the provision of free or reduced-charge services (New York State Senate Research Service, 1990).

• Delivery reform—NYPHRM V authorizes the creation of rural health networks, a new kind of health care institution. These networks will receive enhanced reimbursement of one million dollars each over 2 years and will be given antitrust protection through the so-called “State action” authority. (State action authority permits States to encourage particular types of collaboration and to shield these from Federal antitrust action.) Three such networks are expected to be incorporated during the NYPHRM V period, and it is expected that these will serve as pilots for additional networks in later years. In a related measure, NYPHRM V also authorized planning grants to entities interested in implementing global budgeting demonstrations.

Fifth, and Perhaps Final, NYPHRM

As NYPHRM IV reached its final days, the debate on a successor program centered on a very familiar theme: How to strike the balance among the goals of cost containment, support for hospitals, and expanded access. The trade-offs were clear and quantifiable: squeezing reimbursements to contain costs would hurt hospitals, but greater reimbursements for hospitals would hinder cost-containment efforts and limit the availability of funds for expanding access. Because Child Health Plus and other initiatives introduced in NYPHRM IV were funded out of pool funds, any expansions in these programs would lower the amount available to support hospital charity care efforts.

Overlaying this discussion of the NYPHRM formulas was the beginning of a broader debate concerning the future role of NYPHRM in reform, and a questioning of whether any tinkering with the NYPHRM could get the State to where it needed to be. As part of this broader debate, the Governor's proposal in 1993 called for an expansion of the scope of regulatory activity, including controls on medical fees outside of hospital inpatient settings, and establishment of practice parameters for physicians.

The compromise reached in the final hours of the 1993 legislative session was to approve a 2-year NYPHRM V, which again placed a high priority on support for distressed hospitals. In addition, NYPHRM V continued and accelerated some of the “developmental bank” activities and other delivery reform trends in NYPHRM IV, but made no substantial break with the past in terms of the scope of regulation (Lipman, 1993). For example:

• A new education and training program for primary-care physicians authorized State subsidies for setting up physician practices in underserved communities, incentives for medical students who choose to become family or general practitioners, and grants to encourage academic institutions to develop primary-care training programs.

• A series of new grant and enhanced-rate programs was authorized to help providers develop and experiment with new delivery and financing arrangements, as Rochester did. In addition, NYPHRM authorized grants to help hospitals improve their management information collection capabilities and quality assurance activities, and to facilitate cooperative ventures for shared services. Finally, it enhanced local authority and responsibility for health planning and created a permanent funding source for health systems agencies.

While NYPHRM V was implemented just a few months ago, discussions about a NYPHRM VI or post-NYPHRM world have already begun. This time the issue of how to move to broader reform appears likely to be an even larger part of the debate, and for this reason an important theme in the current dialogue is discussion of the strengths and weaknesses of NYPHRM as a policy tool for achieving future as well as past policy goals.

Evaluation

As previously noted, the original goals for NYPHRM were to curb the growth in health care costs, support distressed hospitals, and improve access by the poor. NYPHRM has been the subject of countless evaluations measuring its success on the cost side, and a few evaluations assessing its performance with regard to the other goals. This review of the literature and discussions with key participants lead to four conclusions about the impact of NYPHRM. First, even among participants who are critical of NYPHRM, there is a sense that ratesetting has served an important purpose for the State. Second, New York has at times had clear, measurable success in meeting one or more of its key objectives. Third, these objectives are potentially contradictory, and the priority ranking among them differs from one time to another and from one player to another. Fourth, the overall level of success one might therefore ascribe to the system will also vary depending on both time frame and political objective.

Cost Control

While the New York system was designed to achieve many different goals, a central concern for the State throughout the decade has been cost containment. Cost has also been the primary focus of the many evaluations of New York and other ratesetting systems (e.g., Gold et al., 1993; Coelen et al., 1988). Discussions with participants and review of these evaluations leads to four conclusions:

1. When containment of inpatient costs has been the primary goal, it has generally been successful. In New York, cost containment has never been the only goal, and at times it has taken a back seat to redistributive goals, but when cost was the dominant objective the State appears to have achieved some success. As in other States, however, the level of success found depends on the time frame, the measure used, and the basis of comparison.

   Evaluation findings have included the following:

   • Levit et al. (1993) found that hospital payments in New York State grew by 9.3 percent a year from 1980 to 1991, below the 9.9 percent growth in the Nation as a whole.
   • Thorpe (1987) found that the costs per adjusted admission in New York State increased by 28 percent from 1982 to 1985, growing at the same rate as the national average, notwithstanding the fact that all payers had a payment surcharge to cover bad debt and charity care. Without this surcharge, payments would have increased by only 23 percent.
   • Hadley and Swartz (1989) estimate that hospital costs under NYPHRM were 13 percent lower in 1983 than they would have been without the all-payer system, and 15 percent lower in 1984.

   According to Thorpe and Phelps (1990), two design features have been particularly important in controlling costs. First, the consistent base year has restrained cost increases which would have been incorporated if a floating base year had been used. (On the other hand, periodic changes in the methodology have included recalculation of the base year allowable costs.) Second, the ratesetting mechanism has disallowed some expenses for above-average-cost hospitals.

2. Containing unit payment increases does not necessarily limit the growth of utilization. Under per diem methodologies, hospitals have no incentive to limit length of stay, and may even have an incentive to increase it, particularly when, as in New York, the base rate is computed on the basis of the very high occupancy levels which prevailed in 1981.

   Even when hospitals have an incentive to limit utilization, factors outside their control can (and to some extent should) limit their ability to do so. In New York's case, the crack epidemic and related public health problems between 1986 and 1987 increased the number of discharges in long-stay diagnoses affecting uninsured or publicly insured patients. As a result, even though hospitals had a strong incentive to reduce the length of stay, average length of stay actually increased starting in 1986 (Vladeck, 1993).

   According to Schramm, Renn, and Biles (1986), hospital expenses per capita were consistently higher from 1972 to 1984 in the group of ratesetting States than in the group of non-ratesetting States, though the differential dropped from 26 percent in 1972 to 11 percent in 1984. In the case of New York, according to Finkler (1987), per capita hospital expenditures grew by 8.3 percent a year between 1983 and 1985, compared with a 4.8-percent national mean.

3. For ratesetting to be a workable incentive for hospitals to reduce their costs (rather than simply a budgeting device for payers) would require at least three things, all of which are weak or absent in the New York case: clear and non-contradictory incentives, hospital freedom to act on them, and hospital ability to control costs through these actions. In New York, however, the incentives are complex, shifting, and in places contradictory; the regulatory atmosphere limits the range of options hospitals can use to limit costs and focuses management decisions on gaming the system rather than reconfiguring health care; and many of the most critical cost factors are driven by changes in public health (e.g., the crack epidemic) or other public policies over which hospitals have little control (Blank, 1993; Miller et al., 1993). Declines in the financial status of already stressed facilities trigger heightened demand to let up on cost containment and focus primary attention on the alternative NYPHRM goal of providing support to distressed hospitals.

4. Systemwide inpatient cost containment has made Medicaid cost containment more difficult. For officials in most States, cost containment largely means Medicaid cost containment. One way States have been able to achieve some control over Medicaid costs is by keeping Medicaid payment levels low (as low as possible without incurring a hospital or nursing home lawsuit under the Boren amendment), and then letting hospitals and others shift costs to private payers.

   In New York State, however, the level of Medicaid inpatient reimbursement has been coupled to that of Blue Cross, and the rates of commercial payers, in turn, are set at a fixed amount above the Medicaid rate.³ So reducing Medicaid reimbursements cuts payments throughout the system. It also cuts the inflow of local and Federal dollars, since each State dollar (including one raised through hospital assessments) is matched by one local dollar and two Federal ones. Given this combination of factors, a Medicaid dollar saved can be ten dollars lost to the State health care system, and therefore Medicaid is seen as a revenue source rather than an expenditure (Fossett, 1993). New York therefore has a strong incentive to expand the Medicaid budget, and indeed has done so, often successfully transferring State-only programs over to Medicaid, where it can receive a match from counties as well as the Federal Government (Fossett, 1993).

On balance, the data suggest that rate regulation, if done in a careful, comprehensive, and politically hard-nosed way, can be a strong tool for restraining growth of unit costs within the regulated sector. While rate regulation cannot affect other important variables driving costs, it can temper the rate of cost increase resulting from these factors. The significant design issue is making sure that the scope of regulation and the unit of payment are most appropriate to the circumstances and political needs of the State.

Support for Distressed Hospitals

As previously noted, a primary goal of rate regulation throughout the decade has been to support distressed hospitals and spread the cost of uncompensated care, and this goal appears to have been at least partially achieved under the all-payer system and other early iterations of NYPHRM. Many hospitals still ran a deficit during these years, but the pools served to redistribute enough funds to reduce substantially both the number of hospitals in deficit and the amount of their shortfall. By 1985, 90 voluntary hospitals in the State were running a deficit, rather than the 152 that would have been in this situation without the pool payments. The combined deficit of these hospitals was less than one-half of what it would have been without these payments—$195 million rather than $467 million (Thorpe, 1987).

During NYHPRM III, however, the policy emphasis focused more on cost containment than stabilization of providers, and hospital revenues suffered. The pools, in the meantime, covered a shrinking portion of the hospitals' bad debt and charity care need, in part because of declines in the assessment rate, increases in bad debt and charity care, and growing tendency to tap the pools for other purposes (Caligiuri, 1993). The result, as documented in an evaluation done for the Council on Health Care Financing (Miller et al., 1993), was a substantial decline in bottom-line margins (the difference between all revenues, whether patient-related or not, and expenses) as well as operating margins (the difference between operating revenues and expenses). Bottom-line losses for the State's hospitals almost doubled between 1987 and 1990, going from $500 million to almost $1 billion. Operating margin deficits also almost doubled during the same period, from $700 million to almost $1.4 billion.

With NYPHRM IV, attention at least temporarily shifted back to stabilization of hospitals. Largely as a result of revised base rate calculations, bottom-line losses went from almost $1 billion in 1990 to $800 million in 1991. Operating margin deficits improved, going from almost $1.4 billion in 1990 to $1.1 billion in 1991. In 1991, the financial position of the State's hospitals was still found to be the weakest of any State in the Nation. On the other hand, it would appear that the financial standing of at least some of these hospitals would have been far worse without NYPHRM (Vladeck, 1993; HCIA, Inc., 1993; Cleverley, 1993).

Access

At least in the early years, the redistributive pools also appear to have enabled New York to achieve its third, related goal of enhancing access to hospital care for the uninsured.⁴ Thorpe (1988) contrasted the rates of hospital admission, hospital days, and outpatient days for the insured and uninsured before and after NYPHRM I and found a dramatic turnaround in access for the insured.

• On the eve of NYPHRM, the number of uninsured was climbing—growing by 5.3 percent from 1980 to 1982—but inpatient and outpatient hospital utilization by the uninsured was declining. As a result, the number of hospital days per 1,000 uninsured declined by 13.4 percent during this period. Analysis identified New York as one of the few ratesetting States in which cost-containment efforts appeared to be impeding access.

• Between 1982 and 1985, in contrast, uninsured hospital admissions grew by 15.4 percent and outpatient visits by 2.0 percent.⁵ The number of hospital days by the uninsured still declined slightly, by 2.3 percent, but this decline was smaller than the decline (8.4 percent) shown by those with insurance coverage during the same period.

Later work by Thorpe and Spencer (1991) suggests that these improvements in access were directly attributable to the pools, and that the pools continued to serve this function in NYPHRM II, leading to 30,000 more adjusted hospital admissions a year than would have occurred without the pools.

While the pools appear to have succeeded in meeting the access goals of NYPHRM, three caveats should be noted:

1. Whether these gains continued after NYPHRM II, as pool money was reallocated to other uses and hospital financial stress increased, is not known.

2. Use of the pools as a vehicle for improving access may not have been the most cost-effective way to achieve this objective. Thorpe (1988) showed that each $10 in pool revenue generated only about $4 worth of additional care to the uninsured between 1982 and 1985. On the other hand, several participants interviewed noted that the rate add-on was the only politically viable way to raise money needed for the uninsured at the time.

3. By focusing attention on hospital care and holding down physician reimbursements for Medicaid, NYPHRM reinforced the institutional bias of the system. This helped perpetuate a system in which the poor lacked access to a regular source of primary care and therefore used hospital emergency departments instead. As discussed later, NYPHRM IV and V have begun moving these pool funds into primary care, but the bulk of the funding still goes to hospitals.

In short, New York's approach to the access issue has been to focus on serving as a safety net for institutions, on the premise that these institutions will then serve as a safety net for individuals. This approach seems to have worked well, at least during the early days of NYPHRM, despite the fact that the link between State policy and access is a two-stage one.

Other Effects

While evaluations of NYPHRM have appropriately focused on assessing the extent to which the program has met its goals, it is also appropriate to examine the extent to which the program might have had unintended policy consequences.

Quality

As with any program that has succeeded in limiting costs, the New York case raises concerns that payment regulations and changing incentives for providers might result in lower quality. So far, there is no evidence to suggest this has happened, but little systematic research has been done. Some research has looked at the impact of rate regulation (in New York and elsewhere) on mortality, but the results are contradictory. Gaumer et al. (1989) looked at 15 States with prospective reimbursement and found that many, including New York, had somewhat higher than predicted mortality rates in 1983 among hospitalized Medicare patients, though there did not appear to be any relationship between the amount of cost savings and the amount of increase in mortality. On the other hand, Smith, McFall, and Pine (1993) examined data on Medicare patients hospitalized in 1986 and found that mortality rates in New York (and several other regulated States as well) were lower than estimated.

System Complexity

Although there have been no formal studies assessing the complexity of the New York ratesetting system, such a study would be superfluous. All parties in New York would stipulate that the system is impossibly complex, with the formulas and methodologies changing and growing more intricate each year. This complexity has very serious political and financial consequences, for two reasons. First, financial incentives do not work if decisionmakers do not understand them. The system reportedly has become so elaborate and arcane that many hospital chief executive officers and other leaders do not understand how it works at all. Second, financial incentives do not work when they are internally inconsistent. A recent evaluation (Miller et al., 1993) for the Council on Health Care Financing pointed to several such contradictions. A related problem noted by the evaluators is that the incentives are targeted to hospitals but physicians often make the critical decisions affecting length of stay—a dilemma New York shares with other States.

Despite substantial support for NYPHRM, the very complexity of the system creates a legitimacy problem. When a reimbursement system is both very complex and frequently changed, and when these changes occur through political channels, the pieces come to be seen as arbitrary, temporary, and politically driven rather than a deliberate and coherent policy instrument, and the attention of regulatees becomes riveted on how to change (or “game”) the system (Blank, 1993).

Forging a Conversion Plan

NYPHRM V expires at the end of 1995, and most of the major actors are already thinking about what will happen then. Some foresee a NYPHRM VI, some see 1995 as the beginning of the post-NYPHRM world, but there appears to be wide consensus that an incremental change or fine-tuning of NYPHRM will not be sufficient to achieve true reform. At present, a Governor's Task Force is analyzing the potential New York impact of national reform, the Governor's Health Care Advisory Board is exploring State options for reform, and most of the major players in the State are thinking about what directions the State should be taking next.

A recurring theme expressed in interviews with participants is that the environment has now changed so dramatically that some of the core NYPHRM apparatus is now outmoded, and New York now needs a “conversion plan” much like the plan required to convert resources from the defense industry because of the new diplomatic environment. This section summarizes some of the new forces influencing the New York regulatory environment, recent actions taken in response, and interviewee thoughts.

New Challenges

Several important aspects of the health care environment have changed since NYPHRM began. Four particularly significant recent events have been the move to outpatient care, the growth of managed care, changing health needs, and changes in Medicare.

Move of Services to Outpatient Care

The movement of care from inpatient to outpatient areas of hospitals has been a dramatic national phenomenon (Fraser et al., 1993), and New York State has experienced this change as well. By 1991, outpatient services constituted 23 percent of revenues in New York community hospitals, close to the 24-percent national figure (American Hospital Association, 1992). In addition, in New York as elsewhere, ambulatory services are growing even more quickly in unregulated or less regulated non-hospital sites, such as ambulatory surgery centers and diagnostic centers.

This movement to the outpatient arena has particular significance for New York, for three related reasons. First, while the State has some regulations on the outpatient side, NYPHRM focuses on inpatient rates, and in particular collects pool money from inpatient revenues. Second, hospitals tend to lose money in the outpatient areas in New York, largely because the uninsured and Medicaid beneficiaries rely so extensively on hospital emergency departments and other outpatient areas for their primary care. This problem appears to be growing with the expansion of ambulatory care. From 1987 to 1989, for example, the net outpatient deficit grew by 70 percent (Caligiuri, 1993). Third, as a result, money allocated from the bad debt and charity care pool is increasingly allocated to balance outpatient deficits. Because hospitals need not (and likely could not) relate exactly how these pool disbursements track to individual services for individual recipients, there is concern in the policy community that hospitals are not accountable to the pool, and furthermore that hospitals have an incentive to game the system by moving their losses to the outpatient side (Miller et al., 1993).

In short, as Vladeck (1993) noted about the consequences of this move to outpatient services, “New York State possesses an extraordinary armamentarium for the regulation of inpatient hospital services, but like the drunk who looks for his keys under the lamppost because that's where the light is, the State's tools may be increasingly less well suited to the tasks at hand.”

Collision Course with Managed Care

As previously noted, NYPHRM III introduced what was then a minor modification having to do with managed care, by permitting managed-care entities to negotiate rates. The subsequent rapid growth of managed-care arrangements now has the potential to unravel the web of cross-subsidies and checks and balances which has held the system together. In theory, the DOH reviews all contracts to make sure that hospitals do not agree to rates that are grossly disadvantageous to them, but, for the most part, has taken the position that hospitals are “consenting adults” who would not enter into a contract which would injure them. Because of this change in the law, managed-care entities are now able to obtain a better price than either Blue Cross or the State's own Medicaid program. As a result, managed-care dollars may no longer be available to cross-subsidize Medicaid, Blue Cross, and self-pay patients. In fact, some managed-care entities may even be receiving cross-subsidies themselves. This scenario is raising increasing concern in New York State, particularly in light of the State's historical suspicion of for-profit health entities. Among hospitals, a particular concern is that most hospitals are not in a position to participate heavily in capitated programs on a risk-bearing basis, in part because of the loss of capital and reserves suffered during NYPHRM III.

New Health Needs

As noted earlier, New York has seen a major change in health needs during the 1980s, because of the interrelated epidemics of crack and other substance abuse, AIDS, and tuberculosis. Costs for mental health and substance abuse services, trauma, and neonatal intensive care have soared. While there is no agreement on how to deal with these needs, it is clear that inpatient ratesetting is an ineffective tool for controlling the costs associated with these needs.

Changes in Medicare

Changes in Medicare reimbursement are having an important impact on the viability of NYPHRM. During the early days of prospective payment, most hospitals had a positive Medicare margin, and New York hospitals fared particularly well. In 1986, for example, New York's PPS operating margin was 18.7 percent (compared with 10.1 percent nationally). In 1987, New York was at 15.2 percent, while the margin for the United States as a whole was down to 5.2 percent (Altman and Garfink, 1990). But because of changes in Federal Medicare policy, the margin for New York has continued to decline, and was expected to average between 3.1 percent and -1.2 percent by 1990 (Altman and Garfink, 1990).

ERISA Litigation

As discussed previously, one factor enabling the system to survive until now has been the State's success in avoiding a successful ERISA challenge, in part because of the protection provided by the Moynihan Amendment to the tax code. But most observers in New York believe this success is very fragile, and that an adverse decision in any one of several pending cases could put the arrangement in jeopardy.

State Responses and New Initiatives

New York has been responding to some of these issues with a variety of new initiatives. As explained earlier, each iteration of the NYPHRM has included more provisions which go beyond ratesetting and the inpatient arena. In particular, the Regional Pilot Projects and Child Health Plus redirected funds to support insurance rather than providers, and several provisions in NYPHRM IV and V focus on restructuring the delivery system by enhancing primary care, encouraging global budget demonstrations, supporting the training of primary care physicians, etc. In addition, on a parallel track, the State has also passed several major pieces of legislation designed to change the financing and delivery of care.

Medicaid Managed Care

Legislation enacted in 1991 requires the State to enroll one-half—about one million—of its Medicaid recipients in managed care by the year 2000. The law also stipulates that commercial health maintenance organizations (HMOs) offering plans to State employees must also enroll Medicaid members. As an additional incentive to HMOs, legislation enacted in 1992 created a 9-percent differential between the Blue Cross/Medicaid rate and the HMO rate—but HMOs enrolling a certain percentage of Medicaid recipients are exempted from the assessment.⁶ By April 1, 1993, 181, 485 Medicaid recipients had been enrolled—a 76-percent increase since April of the previous year (New York State Department of Social Services and New York State Managed Care Advisory Council, 1993).

Private Insurance Market Reform

New York's community-rating law, one of the strongest in the country, prevents insurers from denying coverage or charging higher premiums because of health status or other factors. The New York law requires strict community rating for all individual policies, Medicare supplemental policies, and all policies written for groups of 50 or fewer. Unlike similar laws in Vermont, Maine, and elsewhere, the New York law permits no rate variations (Ladenheim and Markus, 1994).

Given the short implementation time, no systematic evaluation of the impact of this legislation has been done. Preliminary analysis has shown some decline in individual plan enrollment, particularly among healthier enrollees, along with substantial premium increases for those who remain in the plan. These early findings have received considerable national attention, and in particular have been used to bolster the argument that insurance reforms, in the absence of universal access provisions, will exacerbate rather than improve the insurance problem. Case study interviews, as well as a recent analysis by Ladenheim and Markus (1994), suggest that the law's impact is somewhat more murky and complex than these depictions would suggest, in part because the legislation was one of many public policy changes enacted at the same time. If the conclusions reached by these early appraisals bear up under further scrutiny, however, they could provide significant lessons in the perils of certain incremental approaches both at the State and Federal levels.

Outpatient Payment Demonstration

While NYPHRM is still inpatient-focused, New York has been working to develop new outpatient payment methodologies. Under a grant from HCFA in the mid-1980s, the State developed and piloted a classification and reimbursement system for ambulatory care. The system includes two modules: a Products of Ambulatory Surgery (PAS) system for ambulatory surgery and a Products of Ambulatory Care Services system for other ambulatory services. These have been implemented on a voluntary demonstration basis within New York State since 1987. Beginning in 1989, the PAS was used to calculate Medicaid payment for all ambulatory surgery (New York State Department of Health, 1990).

Information System Reform

In an effort to simplify the administrative process of billing and payment, lower hospital receivables, and lower costs, the DOH is developing an electronic claims clearinghouse. With support from the Robert Wood Johnson Foundation, this program is now operating on a demonstration basis, linking 28 hospitals into an electronic claims-processing network. The intent of the initiative, somewhat confusingly labeled the “single payer demonstration program” is to identify and test out technologies which can bring the efficiencies of a single-payer system to New York's multiple-payer system (Ryan, Norman, and Kennedy, 1993). When fully implemented, the demonstration program is expected to save $42 million (Cuomo, 1993). Eventually, the network is expected to include doctors, clinics, pharmacies, and others.

Quality Initiatives

New York also has developed and implemented a variety of innovative programs in the area of quality. Since 1987, the State has been collecting and disseminating information on adverse patient incidents, and has used this incident-reporting system as a basis for developing guidelines on how hospitals can improve their performance through training and experience. A study of the quality of open-heart surgery performed in New York led to the development of software that hospitals can use both to calculate risk of death for any patient prior to open-heart surgery and to compare their outcomes for these procedures with that of other hospitals. As a result, risk-adjusted inpatient mortality rates for these procedures fell by 36 percent (Cuomo, 1993).

A Quality Improvement Initiative proposed by the Governor in 1993 would take State involvement in quality a step further, authorizing the DOH to identify areas for development and implementation of practice guidelines. Under this proposal, the DOH would collect data and then, using a methodology similar to that followed in the coronary artery bypass surgery studies, develop measures of quality of care. These quality measures would be disseminated to hospitals and physicians for their use in improving quality.

The unique and controversial part of the Initiative was the proposal to establish a process for linking these quality measures to reimbursement While the proposal was not spelled out in any detail, the expressed intent was to work with the health care industry to define when it is appropriate to perform specific health services, and find a way to use the products of this research (quality measures, practice guidelines, and assessments of new tests and treatments) to guide medical practice. Some of these uses are fairly common: sharing information with hospitals on best practices, providing opportunities for physicians to learn about how their peers improved outcomes, and providing technical assistance for providers on how to use quality measures to improve care. Somewhat more controversial, however, was the Department's proposal to use study findings to restructure the reimbursement system in order to financially reward care that is more efficient or produces better outcomes. For example, the State might: pay for caesarean births only if patients had characteristics determined as appropriate for surgical intervention; set lower reimbursements for hospitals with higher than standard complication rates; or link reimbursements directly to patient improvement in rehabilitation fields (New York State Department of Health, 1993).

Legacy of NYPHRM

Regardless of the future path of reform, the regulatory experience of NYPHRM and related legislation gives New York some important assets for making and implementing health care policy—including extensive data systems and capabilities, a seasoned and sophisticated group of administrators, and an educated and involved legislature. While some of these strengths will be particularly helpful under a regulatory or hybrid approach, most will be useful in any scenario for the future.

Probably the most significant legacy, however, is New York's history of successful activism on health issues. The choice between regulation and competition is often portrayed as a choice between government and market control, but in reality either approach requires an activist government. While people might disagree about which approach New York should take, or which governmental actors should lead the effort, New York's history of successful involvement in hospital regulation creates the expectation that government should do something, and reinforces the notion that involvement in health care is a legitimate activity for the State.

Next Steps

Like most States, New York is now looking at an array of options for reform, including single-payer models and versions of managed competition. A cross-cutting issue for New York is developing a conversion plan for NYPHRM: Determining where (and whether) the ratesetting system, the sophisticated regulatory apparatus that administers it, and the State's broader regulatory approach will fit in the new system. On this issue, the State basically has three choices. It can:

1. Abandon the regulatory approach.

2. Continue to build on NYPHRM and related regulations incrementally, adding in regulation of physician fees and other outpatient services, and increasing regulatory activity on utilization and quality.

3. Move away from the specifics of the NYPHRM methodology but incorporate elements of the regulatory approach into a broader reform plan.

Given New York's long and relatively successful experience with a regulatory approach, the first option seems unlikely. On the other hand, even among the strongest supporters of NYPHRM, there is some feeling that NYPHRM's incremental, piecemeal approach to ratesetting—termed micromanagement by some—has become too complex and cumbersome, and that adding whole new arenas and levels of rate regulation to encompass other providers and delivery sites would make the system even more Byzantine.

Under these circumstances, one likely scenario for New York might be incorporation of the regulatory approach into a broader model for achieving universal access. Ginsburg and Thorpe (1992), for example, suggest there may be ways in which all-payer ratesetting and competitive strategies could be combined. In the case of New York, one participant's suggestion is that the State might forge a hybrid model in which the impact of managed competition is tempered by regulation. Under a scenario of this sort, the State might still be involved in regulating rates, but the unit of payment would change from an inpatient per case model to an inclusive capitated one. The State might also set clear criteria for health plans, and might continue to play a strong role in assuring that providers serving the poor or other disadvantaged groups would not be hurt under the system. In addition, New York has an interest in global budgeting systems, and NYPHRM V paved the way for further demonstrations of this approach.

While it is not clear which direction New York will take, discussions with participants suggest that most are looking to go beyond, or build upon, NYPHRM in some way. As suggested by Governor Cuomo in his 1993 Special Message to the Legislature, “we can no longer address hospital reimbursement without considering it in the context of system-wide reform. We cannot tackle one part of our agenda this year, another part later. For our reform efforts to work, they must address hospitals as one part of a larger, interdependent system.”

Lessons Learned for Other States

For the past decade, the NYPHRM has been a tool for controlling costs, supporting distressed hospitals, and broadening access. Both formal evaluations and discussions with participants would suggest that while the approach has not generally brought all three at once, it has at least worked to achieve whatever objective was deemed most important at the time. In the process, New York has learned a lot about the design, the politics, and the implementation of the regulatory approach. These lessons might be distilled as follows.

Design

The effectiveness of rate regulation appears to vary with its breadth, becoming most effective when it covers all payers, sites, and services. When the scope narrows, the incentive is for services and costs to move from regulated to unregulated areas. Partial regulation creates an uneven playing field, encouraging gaming of the system and advantaging unregulated sites.

Adoption and Implementation

Hackey (1993) suggests that the “autonomy, capacity and legitimacy of State regulation” is at least as important as the design of the strategy itself in determining success or failure. In New York's case, eight factors were particularly critical, and probably would be essential in any regulatory or non-regulatory approach:

1. Strength and continuity in political leadership.

2. Support by key actors.

3. Political culture supporting activism in health policy.

4. High technical capacity and administrative expertise.

5. Low turnover among administrative personnel.

6. Strong data systems.

7. Success in (or avoidance of) legal challenges.

8. Fit with complementary policies.

Goal Definition

The broadest and perhaps most important lesson of the New York experience is the need to be clear about the goals of policy. For any policy to be pursued consistently over time, major players need to agree on the major policy goals. While consensus at this level can be more difficult politically than agreement on particular measures, lack of consensus can lead over time to a pendulum effect, in which a lot of what might appear to be movement is simply the back and forth swing of policy.