Priority-Setting in Ontario’s Hospitals · Priority-Setting in Ontario’s Hospitals: Research Report March 2002 Queen’s Centre for Health Services and Policy Research The Joint

Priority-Setting in Ontario’s Hospitals:

Research Report

March 2002

Queen’s Centre for Health Services and Policy Research

The Joint Center for Bioethics University of Toronto

The Change Foundation

Priority-Setting in

Ontario’s Hospitals: Research Report

Investigators Craig Jones

Christian Keresztes Ken Macdonald

Doug Martin Peter Singer Hugh Walker

Research Staff Jocelyn Bennett

Helen Coo Verla Fortier

External Research Staff Julie Gilbert Jim Lavery

Queen’s Centre for Health Services and Policy Research

The Joint Center for Bioethics University of Toronto

The Change Foundation

March 2002

Table of Contents

Executive Summary....................................................................................................................ii 1.0 Context for the Study............................................................................................1

2.0 The Request for Proposal (RFP) .........................................................................1

3.0 Research Questions and Key Terms....................................................................2

4.0 Research Review: Priority-Setting in Health Care................................................3

5.0 Document Review...............................................................................................14

6.0 Key Informant Interviews.....................................................................................23

7.0 Provincial Survey of Hospital CEOs....................................................................39

8.0 Case Study: Accountability for Reasonableness.................................................50

9.0 Conclusions and Recommendations...................................................................59 Appendix 1.................................................................................................................................61 References.................................................................................................................................62

Tables Table 1: Hospitals in the key informant interview sample...........................................................24

Table 2: Proportion of respondents reporting significant difficulties at their Hospital.................41

Table 3: Ratings of degree to which factors constrain or facilitate resource allocation..............42

Table 4: Mean ratings and standard deviations, usefulness to resource

Allocation decisions of formal planning methods.......................................................43

Table 5: Mean ratings and standard deviations, usefulness of data and information

Tools to resource allocation decisions.......................................................................44

Table 6: Mean ratings and standard deviations, usefulness of decision procedures

for patient care..........................................................................................................45

Table 7: Mean ratings and standard deviations, influence of external organizations

and groups on resource allocation priorities.............................................................47

Table 8: Mean rating and standard deviation.............................................................................49

Table 9: The four conditions of accountability for reasonableness............................................52 Figures Figure 1: Hospital CEO control chart.........................................................................................46

Figure 2: Distribution of Ratings.................................................................................................49

ii

EXECUTIVE SUMMARY This report presents the findings from a study of priority setting in Ontario’s hospitals, commissioned by The Change Foundation and conducted by the Queen’s Centre for Health Services and Policy Research, the Joint Centre for Bioethics at the University of Toronto and The Change Foundation. The Study Questions

The study questions defined by The Change Foundation were:

1. What resource-allocation priorities do hospital managers have?

2. How do hospital managers make priority-setting decisions?

3. What is the nature of managers’ accountability for priority setting?

The study examined hospital-level planning, policy and operational decisions that determine priority-setting.

The Data Data to answer the study questions were drawn from:

• Interviews with 35 Chief Executive Officers (CEOs), Board Chairs and Chiefs of Staff at 12 hospitals varying in size, region and type of facility;

• A province-wide survey of hospital CEOs, completed by 86 respondents representing

53% of the hospital corporations in Ontario;

• A review of legislation and regulation, policy statements, guidelines, procedures, information and advice used throughout the hospital system;

• A review of research on priority-setting in health care; and,

• One in-depth case study of priority-setting decision-making in a hospital.

A Management Report outlines the justification for the findings and defines central concepts in more detail for readers who need a concise overview of the study results. This Research Report elaborates upon the methodology, explains conclusions in depth and relates the findings of the study to organizational theory, politics and ethics, for those with a scholarly interest in the nature of decision-making in Ontario’s hospitals.

iii

CONCLUSIONS An analysis and synthesis of the data generates the following conclusions:

A. On priority setting Within the framework of priorities established and governed by the Ministry of Health and Long Term Care (MoHLTC), hospital managers select priorities and allocate resources in response to the unique needs of their communities and within the constraints imposed on their operations by their fiscal position, their ability to recruit and retain staff and their ability to raise funds locally.

The recruitment, retention and compensation of professional staff -- particularly nurses, community-based physicians and specialists -- is a challenge facing almost all hospitals and has become a common priority across the system.

Senior hospital managers across the system prefer to give priority to community access to quality care notwithstanding the fiscal challenges. At the service level, however, where specific resource-allocation decisions determine access to and quality of specific services delivered, priorities vary significantly from hospital to hospital.

B. On how priority setting decisions happen All hospitals are required by law to have a minimum organizational structure through which decisions are made and ratified, at least in theory. The actual structures of governance and management at the hospital level vary across the system. Major strategic and operational planning decisions are generally taken through the governance and management structures of the hospital. Directions, resources and approvals from the MoHLTC limit the options available for these types of decisions. The day-to-day interpretation of priorities is generally made by senior managers or through ad hoc consultation between senior managers and executive board members. The dynamics of change -- generated by resource scarcity and community demands -- are faster than can be accommodated by traditional strategic planning processes. Traditional planning is not sufficiently responsive to rapidly changing environmental and operating conditions. Long-term priorities are routinely pre-empted by immediate pressures and contingencies. What might once have been regarded, as "crisis management" has become routine, daily decision-making.

C. On accountability for priority setting Hospital governors and managers have multi-directional and often conflicting accountability; to funders; to patients; to communities, and to the broader public all of who, in one way or another, “confer a responsibility.” Senior hospital managers confront conflicting demands on a continuing basis and face the challenge of negotiating with the MoHLTC and managing the results of those negotiations to satisfy their accountability to their more immediate constituencies.

iv

There was considerable variation in the extent to which hospital decision makers used the various instruments of accountability present in the system. One of the strengths and perhaps also one of the weaknesses of the Ontario system of public hospitals is its rich diversity. The governance and management of a hospital in Ontario is no easy task and decisions on the selection of priorities and the allocation of resources are severely constrained by externally imposed conditions. In Ontario’s hospital system, priority setting is highly contingent upon idiosyncratic combinations of determinants and pressures. This variability, combined with a tradition of relatively high levels of public funding and a relatively low level of micro management by public funding bodies, has supported the development of a system that produces unique and innovative approaches by individual hospitals. That same tradition, however, has arguably contributed to an oversupply of institutions and acute care beds. The creation of the Health Services Restructuring Commission (HSRC) -- and the ensuing system-wide, short-term, rationalization process -- reduced, at least temporarily, the decision-making latitude of many hospitals. Thus Ontario has evolved an acute care hospital system in which:

• Decision-making often occurs outside formal organizational structures and decision-making processes. Hospital managers employ a wide variety of approaches in their decision-making for priority-setting practices but do not see structured processes as particularly significant to the outcomes. This appears appropriate in view of the uncertainty present in the identification of, and solutions to, priority-setting problems.

• Who is most influential in decision-making varies widely among different hospitals, their governors, managers and communities.

• Hospital managers, boards and staff value accountability, see themselves as accountable and generally behave accountably.

• Take-overs and receiverships are rare. Both common and location-specific issues condition -- and sometimes drive -- priority-setting decisions in hospitals in Ontario. Among the common issues facing many hospitals are: the resolution of issues arising from forced mergers and acquisitions, the legacy of financial shortfalls or deficits and protracted conflict between budget management, standards of care and consumer demand for quantity and quality of services. These challenges may be coupled to that of location-specific issues such as serving large catchment areas with low populations -- calling into question the economic viability of the hospital -- which itself gives rise to another issue; that of attracting and retaining clinicians while not burning out the existing staff complement. Some common factors -- the influence on clinical priorities of fee-for-service payment systems, the costs of emerging technologies or the unpredictability of budget approval schedules -- are not within managerial control at the hospital level. Other issues -- like the ethics of clinical priority-setting (Who gets the hip?) or the challenge of creating a continuum of care -- are location- specific but embedded in a larger context; either of society-wide debates over the allocation of scarce resources or, localized contests over the expenditure of community resources. Some hospital managers are hampered in their management by limited data for cost/benefit analyses concerning clinical decisions (Does it make sense to purchase the cheapest hip?) and by under-developed and under-utilized information systems. An additional constraint and complication for hospital managers arises from the difficulty of raising money from small communities or the imposition of

v

provincial planning agendas displacing local priorities. Nonetheless, in the face of such difficulties and system-wide constraints and challenges, many hospital managers succeed at their core missions through resisting the temptation to be all things to all people, by making hard choices about not providing services where there is insufficient demand to ensure continued high levels of skill. Utilization of off-site clinics permits better patient access and lowers overhead operating costs while establishing hospital-run but separate programs for chronic and long term care to better manage patient flow. Hospital managers also understand that when the survival of their institution is at stake -- e.g. when threatened with the appointment of a provincial supervisor -- they may preserve their core mission by relying on highly structured decision-making processes. Diversity and decentralization are strengths of Ontario’s hospital system. Diversity reflects the different circumstances, pressures and constraints characteristic of each setting, region, municipality and hospital management team. Decentralization is required to respond to the diversity of local needs. Decision-making for priority-setting should be, as much as is feasible, matched to local pressures, constraints, needs and circumstances. The adoption of a single decision-making and priority-setting process on a province-wide basis may undermine the functioning of a system that -- by and large -- delivers good quality service in an efficient and accountable manner.

Recommendations on priority-setting research, and its dissemination to, and use by, hospitals were made to The Change Foundation, The Ontario Hospital Association and the MoHLTC.

1

1.0 CONTEXT FOR THE STUDY The hospital system in Ontario is under extreme pressure from the cumulative effects of constraint and cutbacks and from demographic changes in the population that are leading to changing patterns of demand for changing mixes and volumes of services. Hospital managers struggle constantly with determining priorities when allocating human, fiscal and physical resources. Research into the state and processes of decision-making in hospitals in Ontario is therefore timely. Specifically, there is a need to identify the “who” and “how” of decision-making and to identify “best practices,” to provide policy makers, governors, managers, administrators and other decision makers with clear and precise information on the nature of decision-making among their peers.

2.0 THE REQUEST FOR PROPOSAL (RFP) The mandate of The Change Foundation is to promote, support, encourage, and improve health and health care delivery within an Ontario context. “Priority decision-making” has emerged as a central question for health care providers. The Foundation has chosen to address this through contracting a broadly based review of the current state and processes of decision-making in Hospitals in Ontario. Accordingly, the Foundation issued a limited circulation RFP in December 1999 expressing interest in “applied research” to help “identify transfer-to-practice opportunities.” The RFP called for “Exploratory research to describe priority decision-making in Ontario hospitals in order to make more explicit the assumptions, methods and processes and thereafter to compare these findings to models, standards or best practices reported in the literature.” This project, then, seeks “to explore priority decisions related to fiscal, human, and physical resources in terms of what elements various stakeholders deem to be most important, describe how priority decisions are made in hospitals in Ontario and make suggestions as to how decision-making processes can be improved.” The study has two objectives:

1. To describe how decision-making goes on in the real world of hospital operations with a view to identifying best practices, and

2. To place real world decision-making into a research-based theoretical context. 2.1 The deliverables The specific deliverables identified in the RFP were:

1. To describe priority decision-making in Ontario hospitals;

2. To discover whether there are generalizable processes and criteria for any or some kinds of priority decisions in Ontario hospitals; and

3. To consider current priority decision-making.

2

3.0 RESEARCH QUESTIONS AND KEY TERMS Preliminary analysis of the key informant interviews provoked a re-examination of the RFP’s research questions which, following discussions with The Change Foundation, were clarified and refined as follows:

1. What resource-allocation priorities do hospital managers have?

2. How are decisions about resource-allocation priorities made by hospital managers?

3. What is the nature of managers’ accountability for resource-allocation decisions?

The study also required clarification of some of the essential terms used in this report:

Hospital, for the purposes of the present study, is an institutional entity regulated under the Public Hospitals Act (PHA) in which various decision-making authorities allocate care-delivery resources within constraints imposed by fiscal, physical and human limits. Decision-making is the process leading to enacted choices and non-choices -- made by any number of methods -- which reflects the preferences, values and resources available to decision-makers. The definition includes non-decisions which are every bit as determining in the selection of a course of action. Decisions may be the outcome of deliberation over alternatives or the result of imposed circumstances (i.e. the non-choice that imposes itself on a decision-maker). Decision makers are the persons authorized to select between alternatives. They are responsible for aligning available resources to institutional priorities and managing the day-to-day operations of the organization. This report’s focus is on the decision makers at the top of the organization -- hospital managers including the CEO, the Board of Governors and the Medical Chief of Staff (or Chairman of the Medical Advisory Committee -MAC). However, a great deal of decision-making about resource-allocation also occurs at the front-line, by doctors and other health care professionals. Priority-setting, also known as resource allocation and rationing, is the process, explicit or implicit of deciding "who gets what" namely money, staff, supplies and management attention - and the values, preferences, rationales and constraints that underline those decisions. Priority Setting can occur actively through formal or informal decision-making processes, or passively to fulfill the requirements of prior decisions. Priority setting can be the product of priorities established by decision-makers through formal decision processes; or, as the consequence of sudden changes in the operating environment to which decision-makers must respond; or the result of non-choices. Accountability is a complex concept that can be succinctly defined as “an obligation to answer for a responsibility conferred” (Shortt & Macdonald, 1999).

3

4.0 RESEARCH REVIEW: PRIORITY-SETTING IN HEALTH CARE As demand for health care outstrips resources, the need for systematic reflection upon ethics of resource-allocation decision-making in health care becomes more salient. The purpose of the literature review was to assess research contributions to the ethics of, and accountability for, resource-allocation decision-making in health care, and to the ways these decisions get made in health care organizations.

Priority setting in health care is defined as the process by which decisions are made as to how to allocate health services resources ethically (Loughlin, 1996). The need to set priorities arises from the notion that health care is an area where growing demand has to be reconciled with finite resources (Klein, 1998). This suggests that priority setting involves rationing, or the exclusion of certain services (Schwartz, 2000).

Not everyone holds the view that health services resources need to be rationed. It has been suggested that the problem of finite resources could be tackled by reducing wasteful expenditures, raising taxes or shifting money from other programs (Dixon & Welch, 1991). Moreover, some would argue that the rationing debate arises from a market economics perspective of scarcity, and that the real problem lies in this ideological assumption (Loughlin, 1996). Nevertheless, the prevailing notion is that health services resources are finite and that greater efficiency and/or more money will never prove sufficient to accommodate growing demand. Accordingly, there is a need to set priorities in health care.

Priority setting takes place at many levels: at the macro, or governmental, level; at the meso, or institutional, level; and at the micro, or clinical, level (McKneally et al., 1997). The majority of the literature derives from priority-setting at the macro level. Examples of where priority-setting has been undertaken at the macro level include Oregon, which sought to prioritize health services covered by the state's Medicaid program (described in more detail later); New Zealand, where a Core Services Committee was established in 1992 to advise on what services should be funded under the national health care system; the Netherlands, where the 1991 report of the Dunning Committee made suggestions on how to determine priorities in the reformed social insurance system; Sweden, where a Parliamentary Priorities Commission was appointed in 1992 to advise on priority-setting within the health care system; and Britain, where a working party on priority-setting has been established (Ham & Coulter, 2000).

Many of the insights gained from these experiences can be applied to priority setting at the meso, or institutional level, specifically as undertaken by senior management in deciding how much to spend on various programs and services. An overview of some of these insights follows.

4

4.1 Perspectives on priority-setting and ethical theories of distributive justice Disciplines such as law, medicine and economics offer their own perspectives on priority setting. A legal perspective may emphasize universal rights to health care, a medical perspective may champion the use of evidence-based medicine, and an economics perspective may be concerned with cost-effectiveness analysis (Martin & Singer, 2000). Fundamentally however, priority setting is an ethical issue that requires moral reasoning (Singer & Mapa, 1998; Goold, 1996). Accordingly, various ethical theories of distributive justice have guided priority setting (van der Wilt, 1994), including maximum, egalitarianism and utilitarianism (Olsen, 1997). The first of these, maximum, is the theory of maximizing the minimum position (Mooney, 1987). In health care, this relates to considerations of disease severity (Olsen, 1997). Egalitarianism is concerned with equity (Mooney, 1987). Equity in health care is defined either procedurally as the equal treatment of equals (Mooney & Jan, 1997) or on the basis of outcomes, with the goal of equality of health (Mooney, 1987). In contrast, utilitarianism is based on the principle of the greatest good for the greatest number (Mooney, 1987) and usually defended in terms of efficiency, or maximizing the overall gain in utility that results from health care expenditures (van der Wilt, 1994).

4.2 Tools for priority-setting These different theories of distributive justice have guided priority setting by serving as the basis for various methods, or tools, used to set priorities. For example, the principle of equity underlies priority setting based on either need, age or treatment effectiveness. The use of economic evaluation tools for priority setting is driven by the theory of utilitarianism, which is concerned with efficiency. These tools assess the cost per unit benefit of an intervention (Frankel et al., 1994). One widely used economic evaluation tool is cost-utility analysis, in which the cost per Quality Adjusted Life Year (QALY) gained is calculated. The concept of the QALY was developed in the late 1970s by the United States Office of Technology Assessment. QALYs combine data on the effectiveness of procedures with the value placed on the outcome, which is attained from population surveys (Williams, 1985; in Freemantle & Watt, 1994). Their development was meant to produce a measure that combines the benefits of greater quantity of life with enhanced quality of life as a means to compare the impact of different interventions (Maynard, 1994). Interventions can then be ranked from the lowest to the highest cost per QALY as a means to prioritize health services (Nord et al., 1995). Regardless of whether their guiding principle is maximum, egalitarianism or utilitarianism, all the tools used to set priorities have their drawbacks. For example, the use of need as a tool for priority-setting could have the effect of relegating health promotion programs to the bottom of the list (Mooney, 1998). Aside from the moral implications, the use of age as a rationing criterion can lead to elderly people being deprived of cost-effective care (Maynard, 1994). Moreover, data on the effectiveness of many interventions are lacking (Light, 1991; in Freemantle & Watt, 1994).

5

Economic evaluation tools also have many critics. For example, the use of cost-effectiveness analysis as a tool for priority-setting has raised concerns about equity (Ubel et al., 1996). QALYs have been criticized for an inherent bias towards the aged and the disabled or ill. The difficulty of making social choices based on individual utilities, or preferences, has also been highlighted (Donovan & Coast, 1994), since there is no perfect way to aggregate individual preferences into a single social preference (Varian, 1990). Furthermore, one study that asked participants about their utilities for various health states found that these individuals overwhelmingly rejected the rationing choices that were derived from their utility responses, suggesting that cost-utility analysis may not be appropriate for setting policy (Ubel et al., 1996). Perhaps more fundamental than these drawbacks however is the recognition that there probably never will be a perfect technical solution to priority-setting in health care (Freemantle & Watt, 1994). The goals of a public health care system are a complex amalgamation of a number of other goals, not all of which are readily defined. Thus, the use of any one tool to set priorities within such a complex entity may be overly simplistic (Holm, 1998). This complexity is further underscored by Holmstrom (1999), who states that “priority-setting is a complex interaction of various decisions at diverse levels in the organization. There is no self obvious set of ethical principles or scientific tools to determine what decisions we should take at various levels, nor is there an easy or obvious way to determine what decisions we should take at various levels, nor is there an easy or obvious way to resolve the dilemma of the increasing gap between what we can and would like to do on one hand and the resources on the other”.

Priority-setting should be viewed not as a problem but as a dilemma, which means that there is no one correct answer; there may not even be an answer (Holmstrom, 1999). This situation arises because an individual can always legitimately claim that they have a medical need (Daniels & Sabin, 1997). The establishment of priorities that may deny this same individual medical care must then be based on some value system (McKeown et al., 1994), rather than on what is "right" or "wrong." Accordingly, there will always be moral disagreements over the decisions that are made (Daniels & Sabin, 1997). For example, setting priorities based on principles of efficiency may lead to very cost-effective programs that cannot be offered to everyone because of the expense, thereby violating the principle of equity (Singer & Mapa, 1998). The question then becomes whether priority setting is, or can be, a truly rational undertaking (Holm, 1998; Klein, 1993), or whether it is essentially a value-based political process (Freemantle & Watt, 1994). By acknowledging that the process of priority-setting in health care, like organizational decision-making, involves many people, many goals and many values, there may be a shift from the more rational approach advocated by the economist to the awareness that priority-setting is essentially a political process which involves “pluralistic bargaining between different lobbies, modified by shifting political judgements made in the light of changing pressures” (Klein, 1993, p.310; Conrath, 1967).

6

4.3 The process of priority-setting With the recognition that there is no simple solution to priority setting in health care, it is not surprising that more emphasis is being placed on the process itself rather than on the tools that are used (Holm, 1998). Goold (1996) points out that directing effort towards developing a process rather than focusing on specific tools is advantageous in that a process is more flexible and can be used to set new priorities as evidence and social values change. It is worthwhile to note that in the definition of priority setting given earlier the terms “process” and “ethically” are both used. Priority setting is, or should be, an ethical process. The ethical nature of the process of priority setting does not arise from the tools that are used, even though, as discussed above, these are all derived from ethical theories of distributive justice. Rather, “priority-setting” is a term used to describe the transformation of implicit rules into explicit processes for resource-allocation (Lomas, 1997). Thus, explicitness lies at the core of an ethical process of priority setting. This is in contrast to an implicit process, in which physicians are forced to make choices about how to ration care. Although this may be more politically expedient, it may also foster discrimination. An explicit process enables debate about what values should be used to set priorities and also allows them to be universally applied (Dixon & Welch, 1991). Ethics attempts to “make explicit the individual and societal principles that lie behind judgements and positions” (Sawyer & Williams, 1995, p.1410). This suggests than an ethical process of priority setting requires public input (to understand individual and societal principles) and accountability (to make explicit the rationale for decisions). According to Klein (1993), these features will create a situation of open dialogue and debate, and will present an opportunity to explore different values or preferences. This is what will endow priority setting with a rational nature, or one in which good reasons can be given to justify decisions. An ethical process of priority setting then becomes one in which the subsequent decisions may be viewed as legitimate and fair (Daniels & Sabin, 1997).

4.4 Public input Why should the public be consulted about priority setting? Since the Canadian public, through the taxes it pays, really "owns" the health care system, any policies, including priority setting, should reflect the values and interests of the owners, and not the service providers (Chenoy & Carlow, 1993). Public input allows for the incorporation of society's values and interests (Goold, 1996). Although different groups may have large differences in their underlying values (Edgar, 2000), rendering a consensus difficult to achieve, everyone’s’ views are worth hearing because they reflect the different perspectives of those affected by the process (Stronks et al., 1997). Furthermore, by giving the public ownership of the decisions that are made (Lomas, 1997), public input may also increase trust and add legitimacy to decisions that emerge from the priority-setting process (Mullen, 2000).

7

Some concerns have been voiced about involving the public in priority setting. Those who participate may not be a representative sample (Mullen, 2000). Moreover, lay people may not have sufficient knowledge to make decisions around health care (Mullen, 2000). Priorities based on public values could also oppose the principles of equity and equal access (Stronks et al., 1997; Bowling et al., 1993). While these are all legitimate concerns, some of them relate more to the scope of public input, rather than to the fact of public input. The scope of public input encompasses who is involved in priority-setting, as well as the nature of their input. The former can range from individuals who form part of the existing structure of a democratically elected government, to citizens (a grassroots approach), to appointed or elected bodies of decision makers (selected representation) (Goold, 1996). Even the term “citizens” can imply the whole population or a representative sample, or patients, or community representatives (Donovan & Coast, 1994). What is important is that the process at some level is one of democratic deliberation (Daniels & Sabin, 1997). There may be wide variation in the desired extent of involvement as well (Coast, 2000). Many people may prefer to be involved in a consulting role rather than in formulating specific decisions related to priority-setting (Abelson et al., 1995). Accordingly, public input may be most suited to eliciting preferences or values, rather than taking the form of actual decisions that may require medical expertise (Hadorn, 1991). Goold (1996) argues that the latter requires far too much information than most people would be able to assimilate, and suggests that public input take the form of consent to the process, rather than to the actual decisions made. In this regard, the purpose of public input is not to set priorities that result in improved health status, and therefore it may not be fair to evaluate its effectiveness on this basis, as suggested by Charles and DeMaio (1993). Rather, the purpose of public input is to contribute to an ethical process of priority-setting so that the decisions that are reached can be viewed as legitimate and fair.

4.5 Accountability Accountability, the second feature of an ethical priority-setting process, is defined in a number of ways. For instance, it is “...the seeking of input into decisions and policies through consultation and the reporting back of results” (Sawyer & Williams, 1995, p.1411). This definition makes clear the link between public input and accountability, and ensures that individuals retain some control over the priority-setting process (Goold, 1996). Accountability has been further defined as the obligation to give an account (Schafer, 1999). Accountability therefore calls for clarity in the decisions made and the reasons for those decisions (Coast et al., 1996; in Coast, 2000). There are concerns about introducing such explicitness into the process of priority-setting. These concerns include fear of media exposure and litigation (Daniels & Sabin, 1997); that it is not possible to obtain a consensus about principles (Klein, 1993; Coast 2000); that it will cause instability in the health care system (Hunter, 1993; Mechanic, 1995); and that the potential exists for feelings of deprivation and denial (Coast, 2000). However, Daniels and Sabin (1997) assert that accountability is the only way to show that fair decisions are being made and that those who have established a process that is legitimate are making them. Most people show a strong desire to know if their care is being rationed in some way and why, since this would enable them to assess for themselves whether the decision is fair, and enable them to protest the decision if they judged it to be unfair (Coast, 2000).

8

Yet another definition of accountability incorporates the notion of responsibility, in that it constitutes “the responsibility accepted... for promised results” (Dyer, 2000). This meaning is more akin to the definition of public accountability as being "... the obligation to answer publicly for the discharge of responsibilities that affect the public in important ways" (McCandless & Wright, 1993, p.111). These definitions emphasize the need to ensure that public money is spent as efficiently and effectively as possible. This entails reporting not only what is being spent, but why it is being spent and whether it is being spent well (Rayner, 1986). Performance measures are a means of assessing accountability in this regard (Atkinson & McCrindell, 1997). Performance measures for health care must provide information about a broad range of activities. There are two major types of performance measures: balanced scorecards and report cards (Baker et al., 1998/1999). The balanced scorecard is a strategic measurement tool that was developed outside the health care system to help managers evaluate performance, develop change strategies and monitor operations. The health care field traditionally has relied heavily on financial data; by including such things as customer satisfaction, organizational learning, internal business processes and innovation measures, the scope of performance measurement is broadened.

Report cards serve to enhance accountability by offering publicly documented information on the cost and quality of health care services, mainly to people external to the organization to enable them to compare organizations' performances. Various governments and health authorities are developing health care report cards in Canada. In Ontario, the Ontario Hospital Association has awarded a grant to the University of Toronto to develop a hospital-based report card with indicators in the areas of financial measures and conditions; patient perceptions of care; clinical and utilization measures; and systems integration and change measures. As outlined above, an ethical priority-setting process features public input and accountability. The next section will describe the explicit priority-setting process undertaken by the state of Oregon, which incorporated both these features.

4.6 The Oregon priority-setting process In the late 1980s Oregon made the decision to expand Medicaid coverage, the health insurance program for low-income individuals, to a greater number of people. In order to do this within a fixed budget, it passed Senate Bill 27 in 1989, which called for the creation of a Health Services Commission to formulate an explicit process for setting priorities in health services. The bill was the result of public outcry over the withdrawal of coverage by Medicaid for certain organ transplants, a decision that was made with little input from the public or health professionals (Dixon & Welch, 1991). There was the subsequent recognition of a need for an explicit process for priority-setting that involved the public with clear lines of accountability (Kitzhaber, 1993). The Health Services Commission was composed of five primary care physicians, a public health nurse, a social worker and four members of the public. Their mandate was to draw up a list of services ranked according to priority of importance, based on considerations of clinical effectiveness and social values. Panels of physicians rated the clinical effectiveness of approximately 1000 condition-treatment pairs. Public input was sought to integrate social values into the priority list (Kitzhaber, 1993). These values were elicited in three ways. First, public

9

preferences for a variety of health states were obtained by a random telephone survey of over 1000 people. Second, 47 town hall meetings were held across the state to assess the public's values regarding 9 broad service categories. Third, seven public meetings were held to allow input from special interest groups and other concerned individuals regarding specific services (Dixon & Welch, 1991).

A preliminary priority list was released in 1990. There was widespread criticism of some of the rankings: higher priority was given to dental caps than treatment for acute appendicitis (Ubel et al., 1996), while treatment for crooked teeth ranked higher than treatment for Hodgkin's lymphoma (Dixon & Welch, 1991). One reason for these rankings may have been poor effectiveness data (Dixon & Welch, 1991). A further problem may lie in the inconsistency between people's utilities and their rationing choices. Ubel and colleagues (1996) surveyed economics students to measure their utilities for three states of health using one of three methods: analogue scale, standard gamble or time trade-off. They then presented them with rationing scenarios, which were individualized based on their utility responses, such that they would be expected to be indifferent to the two rationing choices. Instead, they found that the participants placed 10 to 100,000 times more value on treating severe illness than had been predicted by their utility responses. When eliciting people's utilities about severe medical conditions the choices are hypothetical (Shickle, 1997). When these choices are placed in a social context people tend to give such conditions much more weight (Ubel et al., 1996).

Some people may view the Oregon priority-setting process as procedurally flawed because of problems with the tools used, which included effectiveness data and utilities. However, because it was an explicit process its elements were transparent and therefore were open to criticism and debate. As stated previously, this explicitness lies at the heart of an ethical process of priority-setting. Another view is that the Oregon process was conceptually flawed because it sought to ration health care; in other words, regardless of how ethical the process may have been, rationing itself is not an ethical undertaking.

4.7 Accountability for reasonableness One framework for an ethical process of priority-setting is described by Daniels and Sabin (1997). Although originally conceived for managed-care organizations, their framework can also be applied to publicly funded systems. They identify two problems in priority-setting:

1. Legitimacy, or under what conditions should authority over priority-setting be in the hands of a particular group?; and

2. Fairness, or when does an individual have sufficient reason to accept as fair particular decisions as determined by priority-setting?

They argue that decisions arising from the priority-setting process can be deemed legitimate and fair it they satisfy four conditions: publicity, relevance, appeals, and enforcement.

These conditions are intertwined with the concepts of public input and accountability. The publicity condition requires that all decisions and their rationales be publicly accessible; in other words, that there be accountability. The relevance condition stipulates that the rationales for decisions should offer a reasonable construal of how the organization seeks to provide "value for

10

money" in meeting the health care needs of the population it serves. A construal will be deemed "reasonable" if it appears relevant to the people affected. Accordingly, it needs to take into account the values and principles of those people. Public input is a means to determine the values a community holds regarding the allocation of health services resources. The two conditions of publicity and relevance therefore fulfill the criteria for an ethical priority-setting process: public input and accountability.

Daniels and Sabin include two more conditions. The publicity condition, because it requires that decisions and their rationales be made public, also creates an environment in which they can be challenged. This leads to the condition of appeals, which requires that dispute resolution procedures be in place that should be dynamic and flexible, and open to review based on new evidence or arguments. A dispute resolution process also provides those affected by the decision an opportunity to re-open deliberations, even if they do not participate directly in the priority-setting process. Thus, this condition can also be seen to incorporate a role for public input. Finally, the enforcement condition stipulates that there be voluntary or public regulation of the process, a condition that enhances accountability. Accountability for reasonableness is a valuable framework for priority-setting in health care. However, it is not clear how an institution might go about operationalizing the model (Gibson et al., 2000). The following section describes how this may be done.

4.8 A Model for priority-setting based on accountability for reasonableness A recent study of two health care organizations has provided insights into how decisions are actually made (Martin et al., 2000). By combining this information with Daniels' and Sabin's framework, the authors identified what they term a ‘diamond’ model of legitimate and fair priority-setting, comprising six elements (Gibson et al., 2000):

1. Institutions: Priority-setting occurs in an institutional context with a mandate to set priorities.

2. People: Groups of people representing a spectrum of interests are involved in priority-setting.

3. Factors: Justifiable factors in priority-setting include benefit, evidence, cost, cost-effectiveness and equity.

4. Reasons: These factors cluster together in specific priority-setting decisions, which are reviewed for consistency with previous decisions.

5. Process: Priority-setting includes procedural safeguards such as transparency, disclosure of conflict of interest, fair access to decision makers, and fair chairing and leadership of the priority-setting team.

6. Appeals: Decisions are open to review based on new evidence or arguments.

The diamond model attempts to operationalize the demands for legitimacy and fairness by stipulating four goals: reasonableness, transparency, responsiveness and accountability. Although they caution that the diamond model was developed in the context of priority-setting for new technologies and therefore the model may not be generalizable to all priority-setting

11

situations, it still may prove a useful guide for health care decision-makers who are faced with the need to set priorities.

4.9 Organizational decision-making processes Research on priority-setting in health care is giving increased attention to the process of organizational decision-making (e.g., Holm, 1998; Martin et al., 2000); i.e. at how resource-allocation problems are actually identified and solved. Yet, this focus on explicit reasons for priority-setting and decision-making appears to have largely overlooked a substantial and relevant body of research and theory about organizational decision-making in management studies, industrial sociology and organizational psychology. This body of research shows organizational decision-making to be an often messy, implicit, intuitive and ill-defined process, which is at variance with the rational explicitness of decision-making advocated by ethicists. Nonetheless, this body of knowledge about the process of organizational decision-making may yet help health care organizations find practical ways to ensure fairness and legitimacy in their resource-allocation decisions. Individuals and groups make organizational decisions; many departments, viewpoints and other interests may be involved. Both the process and the results of decision-making are often risky and uncertain, given that they occur under rapidly changing conditions and in situations characterized by unclear information and conflicts (Daft, 1998). Such conditions, which have characterized the operating environment of hospitals in Ontario over the past decade, have resulted in hospital managers having increasingly to make decisions that are novel, poorly defined and/or non-programmed. Several distinct and differing models of organizational decision-making have evolved over the last fifty years, which can cast light on resource-allocation in health care. The earliest model, one that remains influential, is that all decisions can be seen as risky choices based on extensive evaluation of options and maximization of utility -- in other words, that decision-making is a rational process. The foundation of the rational approach is systematic problem analysis, followed by solution choice and implementation in a logical sequence. Every aspect of a problem is defined, diagnosed and specified; all possible alternatives are identified and evaluated before choosing, implementing and evaluating the best one. This was the basis for the management science approach to organizational decision-making first developed for the military and successfully used by business and government. This quantitative approach works well for programmed decisions -- which are repetitive and for which there are defined procedures -- and less well for non-programmed ones (Daft, 1998). In practice, organizations try to follow the rational model, but can’t live up to it. Organizational decisions are affected by ambiguous goals, a large number of factors internal and external to the organization and the presence of disagreement and conflict about priorities, yet require the support of different people and have to be made in the context of time pressure. All these factors conspire to preclude a systematic analysis of decision-making; people try to be rational but are constrained. The evaluation of options is rarely extensive and almost never exhaustive, so that maximization does not occur. In this “bounded rationality” view, a coalition of decision makers who agree about priorities performs a limited “problemistic” search among known alternatives and chooses the first one that is acceptable using a strategy of “satisficing” rather than

12

maximizing (Daft, 1998; Lindblom, 1954; March, 1994; Perrow, 1986). For programmed problems, the organization can rely on established routines and procedures, while bargaining and conflict resolution are required for non-programmed problems (Daft, 1998). This “Carnegie model” of decision-making -- named after its progenitors at Carnegie-Mellon University -- has led to an emphasis on what and how decision makers are doing and on building agreement through coalitions, rather than on prescriptions for how they should do it. More recent descriptions of organizational decision-making have expanded on the theme of the random, chaotic, intuitive and implicit nature of the process. The incremental model (Mintzberg et al. 1976) emphasizes how a significant organizational decision is often the result of many relatively small choices. As organizations move through decision points they may hit barriers, which require cycling through previous decisions and trying something new. Organizational decisions -- in this model -- do not follow an orderly progression through identification, development and solution. The Carnegie model and the incremental model have also been combined into an integrative description of organizational decision-making known as the “garbage can model” (Cohen, March, & Olsen, 1972), which aims to describe the flow and pattern of multiple decisions in organizations. This model applies under conditions of high uncertainty caused by problematic preferences, poorly defined cause-effect relationships and turnover of participants. When there is high uncertainty, vertical hierarchy and bureaucratic rules tend to be replaced by organized anarchies. Problem identification and solution may not be connected to each other as a linear sequence of steps. Rather, decisions may be the result of four independent streams of events in the organization: problems, potential solutions, participants and choice opportunities. All of these streams flow through the organization, stirred together as if in a garbage can. When the different streams connect, there are several possible consequences: solutions may be proposed when problems don’t exist; choices may be made without solving problems; problems may persist without being solved; and some problems may actually be solved. The garbage can model keeps the whole organization in view, directing attention to the pattern and flow of the many decisions made within it by different individuals and groups. Recent years have seen an emphasis on qualitative research methods to study real-life decision-making. This approach to studying how people make decisions in natural settings has come to be called naturalistic decision-making. People make decisions based on experience. Rather than identifying and weighing alternative courses of action, expert decision makers recognize a situation as similar to one they’ve seen before and choose an action that best fits the situation. Klein (1996) has developed a theory of recognition-primed decision making, which people use to quickly recognize unusual situations and respond to their underlying dynamics, especially under time pressure. Beach (1996, 1997) has developed another approach to decision making called image theory, which sees decisions not as discrete events but as part of the ongoing flow of life. Image theory argues that, in the stream of experience, choices are infrequent. Decisions may not be needed when past experience provides ways, such as organizational policies, of dealing with problems. Decisions are needed mainly when these solutions don’t work. Key influences on how a person makes decisions are (a) the person’s enduring values and beliefs, (b) agenda of goals and (c)

13

plans for attaining those goals. Each of these three influences can be thought of as having an image in the decision maker’s mind. Image theory attempts to describe how people make decisions in steps. First, people narrow their options by screening them through their values, and eliminating unacceptable options. Then, people choose the best options from among the survivors of the screening. In other words, decision makers first consider what is wrong with options, then with what is right, and perform these steps in different ways. Image theory attempts to explain screening and standards for it, choices and their utility within a single theoretical framework (Beach, 1996). Psychological and sociological studies of decision-making in organizations have made significant contributions to understanding processes but appear to have had less influence on the study of priority setting decisions in health care than might be expected.

4.10 Conclusions and Implications for the Study of Decision-Making in Hospitals Although priority-setting concerns the allocation of health services resources, it’s fundamentally an ethical and organizational issue rather than a medical one because it involves questions of fairness (Singer & Mapa, 1998). Although there are a number of tools to guide priority-setting, which are based on ethical theories of distributive justice, relying on tools alone is an overly simplistic approach to the complex and value-laden nature of priority-setting. Ethics has contributed to the use of tools to help when setting priorities, but there is a need to further expand the role of ethics to shape a process for priority-setting (Sawyer & Williams, 1995). An ethical process of priority-setting is explicit; this explicitness arises through the features of public input and accountability. Transforming priority-setting into an explicit process endows it not only with an ethical nature but also with a rational one, since good reasons can be given to justify decisions (Klein, 1993). Ultimately, such a process will enable the decisions that are made to be viewed as fair and legitimate by those affected by the rationing of health services resources. Alternatively, in view of the previous failure of prescriptive approaches to understanding organizational decision making such as priority setting, the elements of a future framework for priority setting in health care institutions may be found in bringing together explicit normative criteria for priority setting decisions, such as accountability for reasonableness (Daniels & Sabin, 1997) and organizational structures and processes of accountability (e.g. Macdonald and Shortt, 1999), along with existing models of decision making processes, such as Beach’s (1996) image theory.

14

5.0 DOCUMENT REVIEW Priority-setting for resource-allocation -- or for that matter decision-making of any kind -- occurs in a context created and re-created through the ongoing exchange between rule-making institutions and persons with rule-taking institutions and persons. Rule-making is that activity or process which establishes “the limits of the possible” for persons and institutions subject to those rules. Legislatures and parliaments are rule-making institutions by virtue of their legitimate authority (i.e. power) to create legislation, which is enacted into law and enforced by the state (Ham & Hill, 1993). The Health Services Restructuring Commission -- a creature of the Government of Ontario -- was a limited-term rule-making institution vested with wide powers to restructure the delivery of health care services in Ontario.

5.1 A structural account of constraints on resource-allocation There is a significant power disparity between rule-makers and rule-takers: the latter are constrained by statutes, enactments, legislation, regulation and law to adhere to the letter if not the spirit of rules created by rule-makers, while rule-makers are constrained -- in the Westminster system of cabinet government such as has evolved in Ontario -- by constitutions, convention, tradition and political expediency. This disparity in power -- which is inter alia a product of the willingness of rule-takers to comply with rule-makers -- gives rise to a structure of authority, power and capacity by which rule-makers condition, constrain and circumscribe the decision-making latitude (or freedom of choice) of rule-takers. So rule-makers create the structure of power within and around which priority-setting for resource-allocation occurs for rule-takers. But why should rule-takers comply? Compliance, after all, implies that the power of the rule-maker is either legitimate or that the rule-maker is able to in some way punish the rule-taker for non-compliance. Institutions, like governments, ministries and hospital organizations, are designed not to eliminate conflict but to regulate it -- and it is through this capacity to constrain social actors that institutions matter. Power is the currency of political exchange. In a deliberately and institutionally bounded relationship -- such as that between the Ministry and hospitals -- power resides in the “capacity to maintain uncertainty, keep others guessing and avoid strong commitments” (Noël, 2001). This encapsulates what hospital managers feel is the nature of their relationship with the Ministry -- and it is the relationship codified in the relevant legislation that defines the relationship between Ministry and hospitals. But why should rule-takers permit themselves to be maintained in uncertainty, constrained in their decision-making latitude and institutionally bounded by legislation? The answer lies in the legitimacy of values shared by rule-makers and rule-takers -- and in the social power of rule-makers to define those values. Institutions, such as hospitals and the Ministry of Health and Long Term Care which funds them, exist in an ideological framework characterized by three overlapping and reinforcing aspects of power which condition, circumscribe and constrain the freedom of action of individual hospital managers. These three aspects of power are, following Scott, regulative, normative and cognitive. The regulative structure consists of rules, laws, governance systems, power systems, protocols and standard procedures. The normative structure consists of values, expectations, regimes or authority systems, habits of conformity or traditions of performance of duty. The cognitive structure

15

consists of categories, typifications, structural identities, performance programs and scripts -- i.e. directions as to ‘how things get done’ (Scott, 1998). There is, in other words, a surplus of compliance enforcement built into the relationship between rule-takers and rule-makers -- underscoring which is the ideological assumption that, in a democratically elected liberal democracy, the rule-maker’s authority is ipso facto legitimate: put another way,

A [rule-maker’s] power is legitimate to the degree that, by virtue of the doctrines and norms by which it is justified, the power-holder can call upon sufficient other centres of power, as reserves in case of need, to make his power effective (Stinchcombe, 1998).

Structures, in this instance social structure, are common and recurring features of human organization. Social structures are patterns of social interaction, which impart regularity and predictability to human affairs. Structures minimize complexity and uncertainty by producing, reproducing and embedding roles, norms and values by which individuals and groups orient their individual and collective behaviour, define their choices and direct their efforts.1 Over time social structures, or ways of doing specific tasks, congeal or stabilize into more or less permanent features -- called institutions or organizations -- which then become elements constituent of and reproducing the social structure itself. Institutions, such as hospital corporations in Ontario,

….consist of cognitive, normative and regulative structures and activities that provide stability and meaning to social behaviour. Various carriers --cultures, traditions, and routines -- transport institutions and they operate at multiple levels of jurisdiction. [They] are multifaceted systems incorporating symbolic systems -- cognitive constructions and normative rules -- and regulative processes carried out through and shaping social behaviour (Scott, 1998).

Once it is understood who makes authoritative rules and who is obliged to follow those rules, much can be accurately predicted about subsequent decision-making on the understanding that professional bureaucracies (such as hospitals) operate by and large as rule-following entities. The rule-following behaviour which arises from, and is embedded in the nature of, recurring interaction between rule-makers and rule-takers gives rise to a social structure characterized by a balance of authority between institutions which leads to a reasonable definition of Ontario’s hospital system as:

….a network of acute- and emergency-care delivery centres (or ‘hospitals’ as defined under Section 93, Regulations of the Public Hospitals Act) permitted to operate under the legal, regulative and statutory authority of the Government of Ontario which is responsible for the “development, co-ordination and maintenance of comprehensive health services and a balanced and integrated system of hospitals, nursing homes, laboratories, ambulances and other health facilities” (Scott, 1998).

16

The persistence over time of this system, coupled to its rule-following behaviour, permits its identification as a social structure characterized by large rule making and smaller rule-taking institutions. These institutions are, respectively, the Government of Ontario acting through the Ministry of Health and Long-Term Care (the primary rule-maker for Ontario’s hospitals and their managers) and the 160 hospital corporations across Ontario (the rule-takers). Institutions in a relationship of hierarchy -- as defined by their rule-making authority or rule-taking function -- are, of course, also and everywhere human creations. Hospitals, in particular, can be comprehended as social expressions of a community-wide ethic or set of values dedicated to a number of different but complementary purposes concerning how the community attempts to care for the well being of its members. As operatives in professional bureaucracies, hospital managers can be understood (more to the point of this report) as clusters of individuals charged with degrees of responsibility for decision-making and priority-setting for resource-allocation and which would, in most cases, correspond or map onto any other similarly organized professional bureaucracy populated by individuals with professional autonomy, diverse skill sets, various kinds of specialized training and levels of bureaucratic authority. It is important, though, not to de-personalise institutions or the structures in which they operate (Scott, 1998). It can be easy to confuse guise with reality: hospitals don’t make decisions on priority-setting and resource-allocation, people do. 5.2 On hospital managers In this report the term “hospital managers” refers to the people who set priorities for resource-allocation; the ‘troika’ of senior management decision-makers of Chief Executive Officer, Board Chair and Chief of the Medical Advisory Committee or President of the Medical Staff. These persons are charged, by virtue of their status within their institutions, as principal decision-makers and resource allocators who are accountable to the Ministry of Health through the Board of Directors/Governors of their hospital corporation. Hospital managers -- see the section on Key Informant Interviews -- constitute the locus of control at the meso level of Ontario’s acute-care hospital system. In the structure of authority and power which governs the behaviour and permits the local allocation of resources, hospital managers -- as rule-takers -- translate into outcomes for their specific institutional settings the rules made for the acute-care system by the authoritative decision-makers (i.e. rule-makers) at the top of the institutional hierarchy in the Ministry of Health and Long-Term Care and the government of the day. A structural account -- such as that presented here -- does not preclude the possibility that some hospital managers will resist having their decision-making latitude constrained by rule-makers in the Ministry. Some will, and some will even prevail against the flux of circumstance and contingencies that characterizes an environment in turbulence. The relations of power, however, as expressed in the differential in power capabilities and articulated in the documents to be considered below, is unambiguous as far as the participants in the structure are concerned. To the extent that rule-takers operate within the legislative parameters established by rule-makers, both reproduce the power relationships that authorize, support and enhance the power of rule-makers. (Rule-takers may contest these parameters but such is the disparity in structural power of the relationship that rule-takers will do so only through appeal processes, procedures and avenues created and authorized by rule-makers).

17

This last point is important because hospital managers, deputy ministers, Vice-Presidents of Finance or Presidents of the Medical Staff are selected, through a variety of techniques, on the basis of training, experience and past performance. Their training, while not uniform, impresses itself upon managers in ways that matter for their self-understanding and for how that self-understanding plays itself out in the performance of decision-making tasks. Hospital managers are, in the current context, highly valued (and well compensated) individuals who are quick studies, with highly developed intuition, leadership skills and acute sensitivity to patterns. They are “experts” in the familiar sense and it is this expertise -- rather it is the process of acquiring, demonstrating and documenting that expertise -- that identifies them as an “epistemic community.” An epistemic community -- as the name suggests -- is a group of people (or a community of practitioners) who share clearly specified “ways of knowing,” and of acquiring, authorizing and legitimating certain forms of knowledge in preference to others. This community aspect arises principally from the values embedded in their educational backgrounds. Though diverse as these may be they embody the forms and structures of a particular “liberal-individualistic” canon which, in the context of their specialized training in business, economics or health administration, lends itself to acceptance of the kinds of rationalities and procedures embedded in the rule-following function required for successful career advancement through an organization like a hospital corporation. These individuals, in addition to being autonomous actors, are also carriers of the values instantiated in them through their participation in and reproduction of the institutions and organizations they serve. The structures in which hospital managers operate, therefore, “are both the result of past actions -- social products -- as well as the context or medium within which ongoing action occurs.” Action, therefore, including the priority-setting and resource-allocation decisions of hospital managers, “operates to produce -- to reproduce (perpetuate) or alter -- structure.” What follows is an account, derived through qualitative analysis, of Ontario’s acute-care hospital system understood as a social network of care-giving institutions characterized by loose coupling and shared meanings that generates compliance, cooperation and conflict (Forster, 1997). Hospital managers, the documents reveal, operate within a context of multiple and overlapping, reinforcing and contradictory constraints on their decision-making latitude, on their ability to innovate, to command resources as needs and population dynamics in their communities evolve, on their ability to meet the missions and mandates of their institutions and on their organization’s ability to deliver the highest quality care. The documents relevant to this analysis -- i.e. for the purposes of analyzing decision-making and resource-allocation in Ontario’s hospitals -- are rule-making: those that frame the enabling legislation which;

(1) Permits the existence and operation of hospitals in Ontario and

(2) Transmits to hospitals across the province the Ministry’s yearly expectations and assumptions for continued funding and operation.

18

5.3 The Public Hospitals Act and the Ministry of Health Act The Public Hospitals Act and the Ministry of Health Act together comprise the overarching legislative authority governing Ontario’s acute-care hospital system. These acts establish in law the identity of rule-makers and rule-takers and entrench the power disparity between them. Taken together, these documents legitimate the Government of Ontario’s political, economic, social and legal authority over hospitals. A structural account of the Ontario hospital system compels an examination of the language by which authority and power relations are specified as that language is embodied in legislation. From this perspective it is clear that the act intends that ultimate power over the disposition -- from incorporation and operation to cessation -- of hospitals in Ontario is vested in the Minister of Health. Indeed, approval of the Minister is required for any building, institution or premises to be operated or used for the purposes of a hospital (s4.2) and the Minister can at any time -- on the basis solely of the Minister’s evaluation of “the public interest” -- suspend or revoke any approval given or deemed to have been given to any hospital. The Minister’s authority encompasses the creation, survival and cessation of any organization operating as a hospital -- the Minister may require repayment of funds, may reduce the amount of financial assistance, may suspend, withhold or terminate payment in whole or in part, may (s6.1) provide specified services, increase or decrease such services and even direct the hospital to cease operating as the Minister “considers it in the public interest to do so.” In practice, of course, the Minister delegates authority for the actual day-to-day operation of the hospital to its Board of Governors, which creates an arms-length relationship between the Ministry and any particular hospital. The Board, in turn, delegates authority -- as envisioned by the Act -- to an administrator, in effect a CEO, who is appointed by the Board according to an approved selection process. So both in practice and in legislation it is the Board of Governors, which is ultimately responsible to the Minister of Health. The Minister’s legislative authority over any given hospital is revealed in the Minister’s prerogative to terminate the operation of a hospital. Additionally, the Minister is authorized by the PHA to “direct the board of a hospital” -- where the minister considers it “in the public interest to do so” -- to provide, cease to provide, to increase or decrease specified services to a specified volume. Finally, in addition to permitting the existence and operation of a hospital, and to holding ultimate sway over the nature and volume of services to be provided, the PHA reserves one final form of authority, a catch-all clause that stipulates that the Minister may “make any other direction related to a hospital that the Minister considers in the public interest.” Finally, the PHA is unambiguous that the Minister monopolizes the power of rule making and compliance definition for the financial operation of a hospital, without which a hospital cannot continue to operate. Section 32(4) states that the Minister may require hospital subsidiaries and hospital foundations to provide financial reports and returns to the Minister and prescribe the accounting principles and rules to be followed in making those financial reports and returns and the manner in which those financial reports and returns are to be provided.

19

The authority of the Minister over hospitals in Ontario is replicated in the Ministry of Health Act and embodied in the following language:

1. It is the function of the Minister and he or she has power to carry out the following duties:

2. To be responsible for the development, co-ordination and maintenance of comprehensive health services and a balanced and integrated system of hospitals, nursing homes, laboratories, ambulances and other health facilities in Ontario.

3. To institute a system for payment of amounts payable under the Health Insurance Act in the form of payment by the Province of all or any part of the annual expenditures of hospitals and health facilities.

4. To establish and operate, alone or in co-operation with one or more persons or organizations, institutes and centres for the training of hospital and health service personnel.

5. To govern the care, treatment and services and facilities therefore provided by hospitals and health facilities and assess the revenues required to provide such care, treatment and services.

The act is clear that legislative authority is vested in the office of the Minister -- that the Minister is authorized to develop, pay for, operate and govern care in hospitals in Ontario. This is in keeping with the history and conventions of the Westminster tradition of Cabinet supremacy in which power flows from the Minister downward such that the Public Hospitals Act and the Ministry of Health Act establish the limits of the possible in the relationship between individual hospitals and the Ministry of Health and Long-Term Care. Thus it appears that the Government of Ontario -- as laid out in legislation -- intends that hospitals operate on a ‘short leash’ over which the Minister exercises substantial control. In a publicly-financed system -- a single-payer system where that single payer in the government of the day -- the legislative (which is to say democratically authorized) power to withhold finances upon which a hospital is dependent to keep its doors open and patients cared for is the ultimate power conferred in the Westminster tradition of cabinet government. The power to dispense operating funds is the power to constrain decision-making where decisions have financial consequences -- it follows that in law and parliamentary practice the primary decision maker and resource allocator is the Minister of Health and Long-Term Care.

5.4 The MOHLTC operating plan requirements document The structure of authority by which hospitals are governed in their priority-setting and resource-allocation decision-making is, as argued above, framed and codified in the legislation by which Ontario’s hospital system is established and then permitted to spend public funds. Once empowered by legislation it is the responsibility of the Ministry of Health and Long-Term Care (hereafter MOHLTC) to monitor and track the disbursement of public funds. As the Minister of Health is accountable to the legislature, so the MOHLTC is accountable to the Minister and -- correspondingly -- so each hospital is accountable to the MOHLTC. The PHA permits the cabinet to appoint an investigator “to investigate and report on the quality of the management

20

and administration of a hospital, the quality of care and treatment of patients ... or any other matter ... “ In more serious cases, for example where the investigator determines that the management and administration of a hospital is no longer in conformity with the acts and regulations, the PHA permits the Minister to “exercise all of the powers of the Board” -- in effect to displace the board and management. These draconian measures are seldom used because the structure of authority is widely acknowledged and largely adhered to. The structure of authority between the MOHLTC and individual hospitals is reproduced through the exchange of reporting instruments -- surveillance devices -- that the MOHLTC uses to monitor and discipline the allocation of resources across the hospital system. Principal among these documents is the one by which hospitals justify their yearly budget request: The operating plan. But hospitals do not craft their yearly operating plan in a vacuum nor do they craft an operating plan according to the fiat of the Board of Directors, CEO or Chief of the Medical Advisory Committee. The operating plan is a creature of the MOHLTC’s institutional desire to -- as much as possible -- gather information, standardize it into uniform categories and employ the information so gathered to track the variation of health care needs and delivery across the province. The Operating Plan Requirements Document is the template by which the MOHLTC governs the development of every hospital’s annual operating plan and quarterly reports. The template outlines what information the final hospital-operating plan will contain, in what formats it will be conveyed and at what level of detail. In the introduction to the 2001/2002 template, the MOHLTC characterizes the Hospital Operating Plan (hereafter HOP) as “a key element in hospital and Ministry accountability.” The approved and submitted HOP becomes the “base for subsequent monitoring, evaluation and remedial action” [by the MOHLTC]. As seen from the MOHLTC, the HOP is “fundamental to the hospital’s ongoing planning activities.” Within the context of local, district and regional needs, every hospital is required to work up a detailed schedule of programs, services, human resource, financial initiatives and requirements funding requirements. The MOHLTC template stipulates the reporting requirements for the HOPs and constitutes a key decision driver for hospital managers. The HOP also lays out the operating assumptions upon which the hospital should expect to, or not to, receive its MOHLTC funding allocation: note the precise order in which the MOHLTC’s priorities are enumerated in Section 1.1 (Introduction and Background) of the template, “hospitals are expected to”:

1. Operate a cost-efficient organization; 2. Pursue cost, service and utilization efficiencies; and 3. Maintain the best quality patient care, all without service gaps or interruption.

The recurrence of “cost” in the first two of three MOHLTC priorities sends a clear signal to the managers charged with formulating the HOP. The rule-maker is telling the rule-taker one message in two different clauses: “contain costs.” Of course this stipulation only makes sense in the context of the rule-maker’s authority to enforce cost containment, which, as argued above, is explicitly stated.

21

Note that the third priority -- “maintain the best quality patient care, all without service gaps or interruption” which is the raison d’être of the acute-care hospital system -- introduces the fundamental contradiction confronting the managers of Ontario’s acute-care hospital system. When push comes to shove, as it recently has with increasing frequency, hospital managers in the key informant sample tended to tilt to the interests of their communities in preference to balancing budgets. Most of the template consists of forms -- fill-in-the-blank spreadsheets -- for which a hospital’s VP Finance is responsible on the basis of year-by-year rolled-over budgets from department and program administrators and heads. The operative section of the Template, however, in which the MOHLTC lays out its rules, is called “Guiding Principles and MOHLTC Assumptions.” These frame the managerial context for understanding the template’s reporting requirements -- and from there the operational expectations -- as seen from the perspective of the MOHLTC. Qualitative analysis of institutional documents presumes that a document’s drafters have settled on the rank ordering of statements (or guiding principles) with a view to signaling the relative importance of one to another. The principles and assumptions direct that operating plans -- without which hospitals will not receive base funding -- must proceed on the basis of an explicit schedule of requirements which can be read as a hierarchy of Ministry priorities (or normative expectations) for the system’s future development, Item 1 of which (for the 2001/2002 template) is a political requirement to fall into line, to buy in to the long-term MOHLTC vision and government strategy, to play the part of rule-taker, vis.:

1. Support the restructuring of the hospital system, 2. Integrating health services and, 3. Enhancing the public accountability of hospitals.

Support restructuring, integrate services and enhance accountability. These are the priorities of the rule-maker and the rule-maker normatively asserts its priorities to rule-takers. It is hard to imagine how the Ministry could send a clearer signal to Ontario’s hospital managers than that embodied in the template’s opening principle. Item 1 of the template, then, succinctly encapsulates the mandate of the Health Services Restructuring Commission (hereafter HSRC). Cooperation of the managers of Ontario’s health care providers is the sine qua non for successful restructuring. The template continues:

Guiding Principle 2: Support greater integration of health services through close collaboration with other hospitals, community agencies, health professionals and others; Note that this second guiding principle essentially recapitulates subsection (ii) of Guiding Principle #1 restating and emphasizing the HSRC’s insistence that “the true interdependency and seamless interaction of a well-functioning health services ‘system’ does not exist ... [and that] in the absence of a coordinated system, continuity of care suffers.” This is, in effect, a requirement to set priorities and allocate resources with a view to the larger complex of health care delivery institutions and agencies proximate to the hospital. While integration and co-ordination of hospitals has been broadly accepted, there is, according to the HSRC’s March 2000 report,

22

“much less consensus on how to achieve vertical co-ordination and integration across the various components of the ‘system’.”

Guiding Principle 3: Incorporate each hospital’s specific 2001/2002 base allocation as outlined in the December 15, 2000 letter from John King, Assistant Deputy Minister for Health Care programs, which illustrated a revised closing base position at that time. As stated in the letter, Ontario hospitals were provided with an over $850 million increase in funding for 2000/01 to ensure their financial viability. Given this unprecedented investment in the hospital sector, the Ministry expects each hospital to maintain program service/stability within the revised base allocation as outlined in Schedule A of the December 15th letter. Hospitals must provide an analysis that sets out, by major expenditure category, the impact of 1%, 2%, and 3% cost increases. Further revenue assumptions must be both fully documented and supportable for review by the Ministry.

Note the language “hospitals must provide” which arises from the MOHLTC’s rule-making authority, as does the requirement that assumptions be documented and supportable for review by the Ministry.

Guiding Principle 4: Plan for designated provincial priority programs where applicable as part of the hospital’s global allocation.

A recurring observation derived from the key informant interviews is the apparent disconnect between MOHLTC requirements -- such as that identified in Principle 4 -- and the provision of funding necessary to realize the MOHLTC’s objectives. The key informant interviews identified many instances of MOHLTC imposition of priority programs, which were expected to be funded out of global budgets which monies were already dedicated to other purposes. The shortfall had to be covered by deficit financing.

Guiding Principle 7: Improve and maintain health services, achieve efficiencies, eliminate unnecessary duplication and ensure hospitals operate within available financial means.

The rule-maker here returns to the raison d’être of the health care delivery system, in the first clause, before reverting to the overarching priority of efficiency and cost containment in the balance of the principle.

Guiding Principle 9: Respond to HSRC directions and to Ministry reforms and initiatives as applicable.

The recurrence of the HSRC recapitulates the substance of guiding principles 1 and 2 -- with a reminder to hospital managers that, local contingencies notwithstanding, the MOHLTC is the primary rule-maker in Ontario’s health care system. These two sets of documents -- the two Acts and the Operating Plan Template -- constitute the primary decision drivers from the perspective of the primary rule-maker. Together with the relevant regulations embodied in the act, as well as within the context of the codes of conduct of professional organizations, accreditation agencies, etc., these texts dictate “how it will be done”

23

and thereby establish the broad parameters of constraint by which each hospital attempts to fulfill its care-delivery mission. The research question posed the following challenge: “How do hospitals make priority-setting and resource-allocation decisions?” On the basis of the documents relevant to decision-making in hospitals the answer must be “with great difficulty.” The balance of power among the institutions that comprise the acute-care hospital system in Ontario gives predominant authority to the legislatively sanctioned authorities to limit the decision-making latitude of hospital managers at the meso level. This is as it should be in a democratically elected cabinet government model such as has evolved in Ontario and in Canada as well. As rule-maker, the Ministry funds on the basis of its appreciation of the relevant variables across the province. As rule-takers, hospital managers manoeuvre with more or less success between the imperatives of compliance with Ministry rules and meeting the needs of their communities and patients. 6.0 KEY INFORMANT INTERVIEWS 6.1 Purpose and focus The purpose of the interviews was exploratory and descriptive: first to identify the kinds of priority decisions that hospitals have to make, and how these decisions are made; then to formulate a description of priority-setting in hospitals inductively derived from participants’ experience. The interviews were aimed at eliciting the everyday experience of senior hospital managers and board members in making priority-setting decisions. The analysis of the interviews was directed at describing the process of setting priorities and interpreting the meaning of priority-setting decisions for participants. The objective was to learn how senior managers perceived the decision-making and priority-setting dynamics that occurred in their own unique environments and how these dynamics shaped resource-allocation. Also of interest was how hospital managers were held accountable, and how hospital managers held themselves accountable, for their decisions. Analysis of the interviews was also intended to inform the design of the provincial survey of hospital CEOs. 6.2 Method: Site and sample selection The criteria for sample selection were to (a) identify appropriate informants and (b) obtain rich information about how priority-setting decisions are made without regard for adequacy of statistical generalization. Accordingly, a purposive sampling approach was applied by combining a selective sampling strategy (Schatzman & Strauss, 1973) for the selection of sites and informants with a theoretical sampling strategy (Strauss, 1987; Strauss & Corbin, 1998) to elicit relevant theoretical concepts from the interviews.

24

In selective sampling, decisions are made in advance of data collection about sites and informants based on established criteria for inclusion. The three hospital management positions with ultimate authority for resource-allocation decisions -- the Chief Executive Officer, the Chief of the medical staff and the Chairman of the Board -- were selected for the interviews. A total of 12 hospital sites were then selected on the basis of the following criteria:

• = Size; • = Location (urban and rural sites); • = Type of facility (as defined under Regulation 92 of the PHA: acute care, complex

continuing care, Academic Health Science Centre, special purpose); • = Region (north, south, east and west); and • = Travel logistics.

All selected sites agreed to participate in the interviews and participants agreed to have their interviews recorded for transcription (the study, including the key informant interviews, was reviewed and approved by the Queen’s University Research Ethics Board). The characteristics of the hospitals in the key informant interview sample are presented in Table 1. Table 1: Hospitals in the key informant interview sample

6.3 Data collection In theoretical sampling, decisions are made during data collection to sample on the basis of concepts that have demonstrated relevance to the theory as it evolves during analysis (Strauss & Corbin, 1990). Thus, over the course of data collection, the interview questions evolved in response to the analysis. The initial focus on identifying the kinds of priority-setting decisions that hospital managers make shifted to a focus on “how decisions get made.” An interview guide in the form of a list of the main topic areas was prepared through discussion with team members, with reference to research on priority-setting in hospitals and organization theory as well as the interviewers’ executive management experience. A letter introducing the purpose of the study, requesting participation for an interview, and explaining the terms of confidentiality was sent to the chief executive officer, chief of staff, and chairman of the board at each of the hospitals selected for the sample.

Location Urban = 10Rural = 2

Type of Facility (under PHA) General/Teaching = 4General > 100 beds = 5General < 100 beds = 2

General Rehabilitation = 1Region North = 3

South Central = 3South West = 3

East = 3

25

The interviews were conducted by a team of two researchers with senior management experience in the public sector. The interviewers used a semi-structured format, with the main topic areas guiding non-standardized questions and extensive probing. Interviews were audiotaped. Consent to record interviews was sought in the initial letter requesting participation, and again at the beginning of each interview session. All participants consented to recording. 6.4 Analysis The tape recordings of the interviews were transcribed verbatim into text files, which were then imported into the Atlas-ti (1997) program for software-supported qualitative analysis. The interview transcripts provided the data for analysis. The transcripts were analyzed using the method of grounded theory (Strauss, 1987; Strauss & Corbin, 1998), which is a structured interpretative procedure to build inductively derived theory -- i.e. grounded in participants’ experience. Grounded theory method consists of the iterative application of various coding techniques to the data. In open coding, events are given conceptual labels and the concepts are classified into categories, for which properties and dimensions are then identified. In axial coding, the categories are further described in terms of the context in which they occur, the action and interaction strategies by which participants handle them, and the consequences of those strategies. Through selective coding, the central aspects of the phenomenon under study are identified. One researcher, with iterative group input, led the analysis. The interviewers were debriefed and their impressions recorded as memos in the analysis. Audiotaping failed in four interviews. The content of these interviews was reconstructed from the extensive hand-written notes of one of the interviewers. Early analysis also provided feedback for modifying the interview guide. 6.5 Results and discussion Analysis of the interview transcripts yielded a rich and multifaceted tapestry of direct experiences, impressions, anecdotes and insights into the realm of decision-making for priority-selection and resource allocation in the Ontario hospital system. The researchers read and re-read the interview transcripts extensively and -- where possible -- followed up with specific managers for more detailed elaboration and explanation. The interview material was thoroughly tri-angulated with the insights derived from the document analysis, follow up interviews with Ministry officials and hospital consultants, the research of academic literature, insights from a consultant CEO and the province-wide survey of current hospital CEOs. Two key insights emerge which are captured in what follows:

1. Large, middle-sized and small hospitals across Ontario confront decision-making for priority-setting in resource-allocation challenges in vastly different -- though in some respects comparable -- contexts; and

2. The differences in these decision-making contexts do not play out the same way in every different hospital setting but are themselves conditioned by and articulated through the management team’s mix of skills, leadership capacity and cohesiveness.

26

The following items, to the end of Section 6, attach substance to the multifaceted tapestry of direct experiences, impressions, anecdotes and insights from the hospital managers interviewed for this study. 6.6 Determinants and pressures A combination of determinants, both external and internal, shapes the culture and climate within a hospital, and defines:

• How hospital managers frame problems and priorities; • How they make decisions about resource-allocation; and • How they select the strategy and actions required to implement decisions.

All of the hospitals in the key informant interview sample had experienced an increase in the complexity of their determinants, both externally and internally. External determinants encompass the main forces in a hospital’s environment that condition the management team’s ability to make flexible decisions. Most important among these are:

• The Health Services Restructuring Commission’s directives; • The Ministry of Health and Long-Term Care; • Other provincial and municipal government organizations; • Other hospitals and organizations in the community; and • Clients, the local public and the media.

Internal determinants encompass the hospital’s operational and policy history, the current state of its financial, human, and physical resources, all of which impact on the management team’s resource-allocation decision-making. Together, external and internal determinants combine to create pressures on decision-making and priority-selection. These pressures, summarized below, exert a significant impact on both the processes of decision-making and the decisions made. Key informants spoke about:

• Pressures related to the history of the operations and management of the hospital: e.g., were there established decision-making processes managed by committed boards? Were finances well managed in the past? Were there adequate personnel, physical, and financial resources? Was there good communication and morale?

• A more complex and interdependent health care system; hospitals that are backed up with long-term care patients are looking for solutions, and working more closely with other hospitals, CCACs, and long-term care facilities.

• A more controversial system rife with well-reasoned offsetting arguments on issues of financing and services.

• A lack of basic data, information, and intelligence to make evidence-based decisions.

• A trend towards program from departmental management organizational structures, leading to significant changes in organizational roles and relationships.

27

• A mix of care that is being changed by technology. Non-invasive and minimally invasive diagnostics and therapeutic procedures are more widely used while invasive techniques now include robotics and computer assisted surgery. “We are just keeping pace with health care in one of the most robust times I can recall.”

• Pressures related to the hospitals’ performance as seen by the Ministry and the government: e.g. were they solvent? Were they seen by the Ministry as well managed? Were they meeting standards? Did they have good local political support?

• A more unpredictable financial situation, compounded by the perception of a lack of timely response from the MOHLTC, and the imposition of its priorities through special initiatives funding.

• Funding of priority programs on the basis of MOHLTC-directed target volumes reduces decision-making latitude in priority-setting decisions at the hospital level.

• The absence of a framework for focused [regional] management, and hence no formal regional integration. The system is centrally managed and leaves it to hospitals to organize themselves within Ministry directions for hospitals to work together.

• Pressures related to the hospital’s relationship with its community: e.g. did the community own it? Was it a source of pride for the community? Did the community see it as meeting the community needs? Was it able to attract necessary personnel? Did it have good relationships with the GPs?

• A mix of patients with more acute needs, because patients with less acute needs have been moved to other forms of care (e.g., outpatient). Reduced lengths of stay have removed low acuity days.

• The impact of the Independent Health Facilities Act, which has allowed private sector health care delivery to take place outside of the hospital. Office diagnostics are increasingly in competition with hospitals.

• The issue of access to, and quantity of available service in response to changing demographics.

• Patients who are more knowledgeable and actively involved in their own care.

• Pressures related to the directives of the HSRC: e.g. were the directives appropriate? Were adequate resources provided? Were institutions involved in mergers, acquisitions and closures cooperative or resistant?

• The HSRC directed hospitals with fundamentally different cultures to amalgamate --

hospital managers had to devise power-sharing formulas among physicians of each of the amalgamating organizations.

• Increasingly demanding Ministry/Government-mandated requirements for self-funding -- hospital managers must raise capital funds to implement changes required by HSRC directives.

28

• Pressures related to on-going management: e.g. was there a balance in the commitment to meeting services with the need to manage a fixed budget? Were there tensions between governance-management-clinical staff? Were communication, planning and decision-making processes and committees working?

• Access to certain services though physician assessment is provided in the absence of public or MOHLTC input on ethical guidelines, e.g. who should get the knee replacement; the active golfer or the elderly person who has difficulty getting around in his/her apartment?

• Counter-productive power balances, such as physician-directed resource-allocation, and OMA control of fee schedules.

• Growing difficulty in recruiting and retaining staff, and especially of doctors and nurses i.e. “We are moving from a time when we had no money and too many staff to a time where we have more money and not enough staff” or “There will be things that we will not be able to buy -- like talent.”

6.7 The Health Services Restructuring Commission For hospital managers in the interview sample, some pressures were temporary; nonetheless, managers were preoccupied with them. Notably, HSRC directives have had a range of enduring and reverberating effects for many hospitals. The HSRC coloured everything: it created over its time uncertainties that persist to this day. At the time of the key informant interviews, the hospital system in Ontario was struggling to adjust to the planning and operational directions of the HSRC. Many hospitals no longer had control over the direction of their strategic planning. The HSRC pre-empted existing plans and redefined their planning directions. The new directions created an exciting and positive challenge for some hospital managers, provoking active or passive resistance from others. The dynamics of priority selection and resource-allocation evident in the key informant interviews may very well have been conditioned by the immediacy of the HSRC directions rather than the pre-commission dynamics or the ones that might emerge as the system reconfigures itself in the future.

6.8 Organizational autonomy The ways in which a management team responded to pressures in terms of decision-making and priority selection was shaped by the specific configuration of the pressures they faced, and by the leadership, management and governance styles of the managers themselves. The combination of kinds of pressures on decision-making and the modes/styles of response can be used to categorize the hospitals into groups according to their autonomy status. The relationship between pressures and autonomy status appears to depend on: (a) how well the hospital is able to control its environment; (b) how well the hospital is able to buffer its technical core of medical services and professional staff from environmental uncertainty; (c) whether the

29

HSRC’s effects were a constraint or an opportunity; and (d) changes pre- to post-HSRC. Accordingly, hospitals in the key informant survey can be categorized as falling into one of five types of autonomy status:

• Robust: Hospitals that are, and have been, strong, self-sufficient, fiscally sound and relatively stable, able to adapt and carry on.

• Enabled: Hospitals that benefited by HSRC directions and took advantage of them to resolve problems and forge positive new directions.

• Constrained: Hospitals that they were weakened by HSRC directions that required them to compromise their strengths in order to accept responsibility for weaker hospitals in their communities.

• Subsumed: Hospitals with a history of difficulties that were forced into mergers and other changes that they didn’t want.

• Compromised: Hospitals that were inherently weak and unable to change their circumstances.

Key informants’ statements clearly indicated that the directions (or lack thereof) of the HSRC were a defining moment for most hospitals in Ontario. Long-term problems of individual hospital organizations were either resolved or worsened by the decisions of the Commission. In many cases, routine planning processes and long term strategic plans were pre-empted, disrupted or short-circuited by the short-term intensive impact of the Commission. In a very real way, the pace of change in the technologies of patient care is far surpassing the pace of change in the technologies of governance and management. Long-held values, traditions and processes are no longer valid. In neither the low or high autonomy status hospitals were managers and governors comfortable with their clinical service priorities, their patient and financial data, their ability to forecast changing needs, changing service demands and changing resources. Rarely did hospital managers and governors state that they had met the challenges of change and that they had their hospitals on the ‘right course.’

6.9 Unique organizational characteristics of hospitals Hospitals are not organizational entities whose operations, structures, and governance are typical of other government funded, voluntary corporations. They differ in that they are professional bureaucracies. Their key personnel -- physicians and nurses -- have unique relationships with their corporations. Physicians, for the most part, are not employees of the hospital while nurses bargain collectively on a province-wide basis. 6.10 Physicians Physician decisions -- and their resource-allocation consequences -- constitute much of the priority-setting in hospitals. Yet, physician incomes (other than for those few who are on salary and for some who are provided supplementary income by the hospitals) are neither paid nor controlled by hospitals, but through fee-for-service (FFS) billings. Physician costing rates are not set by the hospital but through the FFS fee schedule controlled by the Ontario Medical Association. Physician accountability is multi-directional, less focused on the hospital’s

30

institutional needs and more focused on the patient, professional ethics and professional governing bodies. This arms-length relationship between hospital managers and physicians has a significant impact on the hospitals’ ability to establish priorities and allocate resources. Physicians make their treatment decisions and utilize treatment resources according to their assessment of patient need. Planning for the whole hospital is often complicated by competition between patient need and physician competition for resources. The dynamics of changing patient need and the urgency of specific patient circumstances are difficult to forecast and can easily pre-empt established hospital priorities and redirect established hospital allocations. As key informants pointed out,

The number one decision driver in the hospital is the physician. You don’t get into a hospital unless a physician says you get in. You don’t get discharged until a physician says you are discharged. Nothing is done in the hospital except under a written order of a physician.

The ultimate decision-making process begins and ends with reconciling the conflicting demands of physicians that are on staff and the overall resources of the institution. While doctors are not conscious managers, they clearly create an important force with which hospital managers must contend. This was most evident with respect to priority-setting decisions concerning operating room (OR) time. There was little if any explicit external input into clinical resource-allocation representing differing valuation of procedures. Thus, there were disagreements among medical divisions about the extent to which different disciplines do valuable work. Within each discipline, fee-for-service (FFS) reimbursement for medical services had a profound effect on medical resource-allocation decisions, that administrators perceived themselves as having little control over. Physicians were perceived by administrators as “doing what provides good earnings quickly.” The end result was that requirements to reduce operating room time were addressed by across-the-board cuts that, in the words of one respondent, did not differentiate bunions from breast cancer. 6.11 Nurses Unlike physicians, nurses are generally employees of the hospital in which they practice. However, most hospitals, as individual and independent corporations, have only limited influence on controlling and prioritizing nursing costs. Salary levels for nurses are set through an arms-length bargaining process. The Ministry sets the hospital’s core operating budget. With externally controlled budgets and no control over salary levels, hospital managers are left with priority options that focus on service-level provision even though they have no control over service level demand. Hospital managers exercise control over deployment, scheduling, and workload. The accountability of nurses to hospitals is somewhat like that of physicians in that they have primary ethical and professional accountability to their professional bodies and patients.

31

6.12 Structures and processes The key informant interviews did not identify clear and common patterns of formally prescribed decision-making and priority selection processes. Organizational structures and processes appeared to be almost incidental to executive decision-making about priorities. In general, CEOs gave the impression that they themselves were in charge and that structures, processes and broad participation in decision-making were somewhat incidental to priority-setting. Typically, in the sample, the CEO’s role was to respond outward to the environment and downward to the organization, through board demands for performance from management. The role of the CEO varied by hospital and by hospital pressures. Many Chiefs of Staff and Medical Advisory Committees chairs felt less a part of management and more like “I’m here to advocate to management on behalf of clinical staff and patients.” In being fully occupied by medical management issues, the question arises as to whether chiefs of staffs and MAC chairmen are (a) equal partners in the senior management team or (b) only called upon to resolve operational problems. Most hospitals have converted to, or are in the process of converting from, traditional departmental management structures to program management structures. Executives assume that staff participation in decision-making occurs mainly at the level of program management and that staff influence is conveyed through Chiefs of Staff, Chairs of MACs, and other legislated positions on the board. Some respondents expressed strong positive support for regional direction from the MOHLTC.

6.13 Values The values of managers, and the symbolic image of the hospital, as caring and responsive to the community -- and as effectively managed to the Ministry of Health -- can influence priority-setting, decision-making, and management structures and processes. Values are standards of desirability, goodness and beauty that serve as broad guidelines for behaviour. A shared value system can be identified among the key informants. For many of the interview respondents, an important value was that of the hospital as a public trust, a community resource, for which everything must be done to ensure its survival and, preferably, expansion of the organization’s mandate. This meant not participating in a hospital’s failure or a merger into something else. Thus, for the interview participants, the ultimate goal of priority-setting was the sustainability of the hospital organization, through:

• Recruiting and retaining medical staff to attract money to the hospital, e.g. for academic health centres, to recruit for roles on the service and academic side;

• Being competitive: developing a “corporate infrastructure,” advancing information technology, ‘branding’ the hospital, developing programs, as “creating a magnet organization for the attraction of clinical and other staff” -- “attracting the best people”;

32

• Seizing opportunities for synergies and shared resources: e.g., diagnostics, communications, ambulance, and core care. “Seeing that opportunities do not get left on the table”;

• Aligning with MOHLTC goals to get funds, e.g. program funding; and

• Balancing competing interests/priorities -- “If we do this, we can’t do that.” For many of the key informants, successful management was registered according to the organization’s ability to meet or exceed its mandated mission to provide the best possible patient care within existing constraints and, where possible, to expand provision of that care.

6.14 Symbols A symbol is anything that transmits a particular meaning recognized by people who share a culture. The analysis suggests that governance and management are keenly aware of, manage and respond to, the symbolic messages that a hospital generates and conveys. The key informants were conscious that their hospital organizations held symbolic importance for the communities they served and that these symbols required careful cultivation of the corporate image of the hospital itself. Canadians feel strong attachment to their publicly funded health care system, at the centre of which is the hospital (Compas Inc., 2001). Hospitals have come to be seen as the symbolic expression of how a community exercises its intention and desire communally to care for its members. A hospital has to be seen by its community as a place to be trusted, a valuable community asset to be defended and to be held accountable to the public trust -- all the more so if (as is the case in many rural areas) it may be the only such institution in the region.

6.15 Images Closely associated with values and symbols are images: the representations projected outward from the institution itself. Hospital managers revealed themselves to be conscientious stewards of the symbolic importance and “brand” image of the institutions in their charge. Managers identify themselves as protectors and promoters of the images associated with their institutions -- images which seek to advance the institutional mission while simultaneously solidifying their symbolic centrality to the care mandate of their communities. Hospitals organizations differ in the degree or relative lack of value conflict between incumbents in senior management positions. A lack of consistency between the values of incumbents in governance and senior management positions was associated with conflict among them. Many respondents made statements indicating that they placed high value on staff empowerment. These statements referred to informing staff and giving them a voice in hospital management, e.g., through the mandated Financial Advisory Committee (FAC), although some respondents indicated that they did not get useful input from, or did not value, this source. There were comparatively few references to teams, which was surprising in view of the widespread adoption of program management.

33

6.16 Priority-setting for resource-allocation The result of the rather unique relationships between hospital corporations, their key staff, and hospital circumstances, impacts directly on managers’ ability to prioritize allocations and on the nature of their accountability:

• Hospitals are constrained in their ability to set priorities and to allocate resources.

• The choices available to hospitals in the selection of their priorities are limited by external factors.

• The most common pressure facing hospitals internally is the conflict between fiscal management and service provision. Among some senior hospital officials, there is a lack of congruence of commitment between defined and selected priorities.

• Hospitals attempt to establish clearly defined objectives, design well- established structures, utilize well-designed strategies, and collect viable data. Quite often, however, their attempts are pre-empted by externally imposed variables, including inability to secure necessary staff, MOHLTC-imposed priorities, variable service demands.

• There are areas of priority-setting that are not strongly correlated with resource-allocation. Hospitals may, for example, prioritize “research” or “staff morale” in their objectives, both of which have resource implications that are outside the framework of their controllable budget.

• Hospital managers generally do feel accountable to their consumers and their community, but their ability to be so accountable is mitigated by externally imposed priorities. In many cases they are put in the position of saying, “We would have liked to meet your needs but our prioritization of resources is subject to imposed decisions that are beyond our control.”

• Constraints on authority and resources also limit the extent of hospital managers’ accountability for the validity of their decisions because the range of options available to them is limited.

6.17 Accountability Priority-setting and resource-allocation involve significant decisions for which hospitals are held accountable -- by law, by funding bodies, by customers and by communities. Decision makers are conscious that they are accountable for the quality of their decisions. Accordingly, the concept of accountability was examined as it applies to priority selection and resource-allocation in hospitals. The key informant interviews were also analyzed using the elements of a framework developed in earlier work on accountability in the Canadian health system (Shortt & Macdonald, 1999). This framework identified the processes and tools used to both effect and demonstrate accountability in the health care system. Although the framework was developed to describe accountability as it applies to federal, provincial, and local accountability relationships, it provided some guidance as to what to pursue in the interviews, particularly on meso-level accountability processes in hospitals.

34

6.18 Citizen involvement In our democratic system there is an expectation that citizens have access to membership in, and information about, a publicly financed charitable corporation. A fully extended model of governance for such corporations would include an open membership system from which a Board would be elected and from which an executive is appointed. Governance meetings would be open, at least to the membership, and in camera meetings would be restricted to discussions related to personnel, property acquisition and legal matters. The Public Hospitals Act in Ontario mandates a degree of citizen involvement in hospital boards but does not require open memberships or access to boards and committees. There were wide variations in approach among hospital managers in the sample to the formation of governance, to the involvement of the public and to the extent of external communications. Boards were of varying size and composition and some had unpopular structures imposed by the HSRC. Hospital managers struck the committees required by the Public Hospitals Act and they seemed at least to favour as few standing committees as possible beyond that. The pattern of standing committees (not required by law) and ad hoc committees (struck for specific purposes) varied from hospital to hospital. The interviews produced very infrequent references to committees other than those standing or ad hoc committees struck for strategic planning and priority-setting. A common theme among managers was streamlining boards by reducing members and shifting from an emphasis on community representation toward a preference for business skills. Executives did not see broad membership and participation by community members as particularly important other than for the threat that it implied for the smooth functioning of the board. 6.19 Political activity In earlier work, this component of accountability was applied to elected provincial and federal governments where the accountability of the respective legislatures is quite clear. Hospital Boards are a form of government engaged in the process of governance. Their formation and activities are, however, much less transparent (see above). Accountability in the hospital system becomes much more a function of the acceptance of accountability by the individual governor or administrator. In general, hospital managers have a clear sense of accountability to their Board Chairs (although some hospital administrators and Board Chairs clearly demonstrated accountability role reversals). However, in general, the perceived direction of corporate accountability -- as contrasted from ‘felt’ accountability -- was more towards the primary funding bodies and to “customers” than it was to the public or community. Hospital managers vary in their approaches to the use of political activity to achieve their objectives. Some had frequent and on-going communication with their Ministers of Provincial Parliament (MPPs) while others seemed to ignore the political system in favour of dealing directly with the MOHLTC. Only rarely did the subject of regional or municipal government emerge as a consideration in accountability.

35

6.20 Constitutional practice Hospital organizations are created under the Public Hospitals Act. The Act’s regulations define what they can do and what they are required to do. The Minister of Health is accountable to the legislature and to the Public Accounts Committee for the conduct of the Act and for financial transfers to hospital organizations. Given this constitutional basis for their very existence, responses from key informants suggest that hospitals in general are required to focus their priorities and allocations within the framework required by provincial directions. Key informants described ministry priorities (sometimes perceived as ad hoc) as tending to pre-empt hospital strategic planning. A number of ministry priorities (e.g. emergency, cardiac and cancer care) have associated special purpose directed funding, thus further limiting flexibility in planning and priority-setting. 6.21 Provision of information Hospital managers are required to provide annual operating plans and budgets to the Ministry as a condition of funding. These plans are reviewed by Ministry officials and are either approved or sent back for revision. Through this process, hospital priority-setting and resource-allocation decisions are sanctioned and given approval to proceed. Hospital administrators are also required to provide variance reports to the Ministry as a check against their operating plans. In addition, as incorporated public organizations, all hospitals are required to produce and make freely available an annual report and audited financial statement. Most hospitals develop mid- and long-term strategic plans, which guide the development of their annual operating plans. The key informant interviews, however, revealed considerable variability in commitment to these normal instruments of governance, decision-making and accountability. Although some hospital managers were firmly committed to these processes and used them to guide decision-making, others were less enthused and placed less value on them. The role and importance of strategic planning varied across the sites interviewed. Some managers saw strategic planning as a very useful tool to reveal what they should do, and how. Some appeared to engage in strategic planning because it is fashionable, while others did nothing at all. There was only one hospital in the sample where strategic planning was central, apparently because of the CEO’s commitment to it, board buy-in, and organizational memory of the consequences of its absence. Even in this institution, the relevance of strategic planning to the dynamics of the hospital was questionable. This hospital organization was in control of its pressures, and hence in a position of maximum choice about what the management team could do, such that they could engage in strategic planning. By contrast, at another hospital, with an enduring legacy of financial planning difficulties, strategic planning was a survival mechanism dictated by external and internal pressures.

36

6.22 Delegated activity From an accountability perspective, the limitations on decision options create a limitation on the extent of hospital accountability. By imposing such limitations through self-governing professions, the accountability of the hospital governing board is transferred to the self-governing professional bodies. The influence of the professional boards goes beyond fiscal management and extends to issues of professional rights and responsibilities for patient care. The Public Hospitals Act itself imposes some rights and responsibilities of health care, including a requirement for representation on the hospital board of governors. The Act also requires separately established Medical Advisory Committees, which become a sub-set of governance with extraordinary influence on the hospital board and on management decisions made on behalf of the board. A prime example, often cited by key informants, was the allocation of operating room time. This very significant resource-allocation decision tends to be controlled by physicians and surgeons associated with, but not employees of, the hospital. Related resource-allocation decisions may be guided less by the plans and priorities of the hospital board and more by the needs of professional staff.

6.23 Review functions The Provincial Auditor (PA) in Ontario has comprehensive auditing responsibility for the programs and expenditures of the MoHLTC but does not have the authority to perform either program or financial audits on individual hospitals. Thus individual hospital organizations tend not to be subject to the publicity normally associated with the PA’s annual report. The MoHLTC is, however, subject to the PA’s review and is required to impose the standards of a provincial audit on individual organizations. A key audit standard is good fiscal management. Within this context, managers face a dilemma in their priority-setting and resource-allocation decisions. Sound fiscal management is often equated with a balanced operating budget that can only be achieved through efficiencies and service level control. Managers claim to be facing increased demands for service while at the same time are frustrated by a lack of discharge resources, which confounds their efficiencies. The key informant interviews indicated that individual management teams set radically different priorities in response to the conflict between balanced budgets and the provision of adequate services. Some focus on expenditure control while others are prepared to risk debt and deficit in favour of meeting service demands. The direction taken by each management team conditions their priority-setting and resource-allocation decisions.

37

6.24 Managerial functions In the interviews, neither managers nor governors spontaneously highlighted their use of management tools, or of comparative indicators like benchmarking and hospital report cards. They tended, instead, to focus more on the immediate pressures that imposed upon their decision-making i.e. the variables that required diversion from their strategic objectives. This suggests that managers experience a lack of control over their own priority selection and allocation decisions. One key informant pointed out that hospitals across the system are moving from departmental management to program management and that this movement has accelerated despite the lack of evaluation on the relative merits of program vs. departmental management.

6.25 Legal contracts Contracted services are generally seen as a routine approach to achieving operational objectives. At least one hospital has developed a unique relationship with construction contractors who have agreed to finance front-end expenditures on building projects pending Ministry capital approvals. This is seen as a strategic initiative strongly associated with priority-setting and resource-allocation. 6.26 Accreditation and credentialing Many key informants indicated that they took the accreditation process seriously and that they incorporated the accreditation recommendations into strategic planning, priority-setting and resource-allocation decision processes. In some cases, accreditation became the primary tool for setting strategic and operating plans. 6.27 Complaints procedures Hospital managers and governors did not raise the issue of formalized complaints or appeals procedures as a function of their priority-setting and resource-allocation decisions.

6.28 Ethics Ethical behavior is very much on the minds of governors and administrators as they select priorities and determine resource-allocation. Very difficult allocation decisions have often to be made without guidance -- e.g. “Do I give a new hip to the fifty-five year old man who wants to continue golfing or to the eighty-two year old widow who wants to remain self-sufficient in her own home?” Managers argued that guidelines for these types of decisions should be through some mechanism created and sanctioned by society at large. This simple illustration suggests another critical influence on priority selection, resource-allocation, and accountability; the smaller the institution, the closer to its community, the more accountable its managers felt themselves to be.

38

6.29 Conclusions Within the framework of priorities established and governed by the MoHLTC, hospital managers select priorities and allocate resources in response to the unique needs of their communities and within the constraints imposed on their operations by their fiscal position, their ability to recruit and retain staff and their ability to raise funds locally. The day-to-day interpretation of priorities is generally made by senior managers or through ad hoc consultation between senior managers and executive board members. Senior hospital managers across the system prefer to give priority to community access to quality care notwithstanding the fiscal challenges. At the service level, however, where specific resource-allocation decisions determine access to and quality of specific services delivered, priorities vary significantly from hospital to hospital. The dynamics of change -- generated by resource scarcity and community demands -- are faster than can be accommodated by traditional strategic planning processes. Traditional planning is not sufficiently responsive to rapidly-changing environmental and operating conditions. Long-term priorities are routinely pre-empted by immediate pressures and contingencies. What might once have been regarded as "crisis management" has become routine, daily decision-making. Hospital governors and managers have multi-directional and often conflicting accountability; to funders; to patients; to communities, and to the broader public all of who, in one way or another, “confer a responsibility.” Senior hospital managers confront conflicting demands on a continuing basis and face the challenge of negotiating with the MoHLTC and managing the results of those negotiations to satisfy their accountability to their more immediate constituencies. There was considerable variation in the extent to which hospital decision makers used the various instruments of accountability present in the system.

One of the strengths and perhaps also one of the weaknesses of the Ontario system of public hospitals is its rich diversity. In Ontario’s hospital system, priority-setting is highly contingent upon idiosyncratic combinations of determinants and pressures. This variability, combined with a tradition of relatively high levels of public funding and a relatively low level of micro management by public funding bodies, has supported the development of a system that produces unique and innovative approaches by individual hospitals. That same tradition, however, has arguably contributed to an oversupply of institutions and acute care beds. The system-wide, short-term, rationalization process by Health Services Restructuring Commission reduced, at least temporarily, the decision-making latitude of many hospitals.

39

7.0 PROVINCIAL SURVEY OF HOSPITAL CEOs 7.1 Purpose The purpose of the survey was threefold:

1. To assess the generalizability of selected concepts identified in the key informant interviews;

2. To assess new questions arising out of the key informant interviews; and

3. To identify possible sites for the case studies. Specifically, the five main questions addressed by the survey were:

1. What determinants and pressures affect priority-setting in hospitals across the province?

2. What are the main kinds of priorities pursued by hospitals across the province?

3. What is the perceived usefulness of various procedures and practices for resource- allocation decision-making?

4. What influence do various position levels in the hospital and various external organizations have on resource-allocation priorities?

5. To what extent do respondents say they apply accountability for reasonableness criteria in their priority-setting decisions?

7.2 Method Based on analysis of the key informant interviews, an 18-item questionnaire was drafted. The questionnaire asked manager-respondents to:

• Provide selected background information on their hospital, and indicate whether their hospital was experiencing significant difficulties in a number of areas;

• Rate the effect of a variety of factors on resource-allocation;

• Rate the usefulness of various formal procedures and tools for resource-allocation decision-making;

• Rate the influence of various position levels and external organizations on resource-allocation priorities (based on Tannenbaum, 1968; Katz & Kahn, 1978); and

• Rate their hospitals priority-setting in terms of Daniels’ & Sabin’s (1997) accountability for reasonableness criteria.

The survey was pre-tested on three senior hospital managers, which resulted in improvements to the wording of items.

40

Along with a cover letter, the survey was faxed to CEOs of every hospital corporation in the province. One hundred-sixty surveys were faxed out. Respondents were asked to fax back their surveys or complete a web-based version.

7.3 Results and discussion A total of 86 completed surveys were received for a response rate of 53%. Thus, caution must be exercised in generalizing findings from the survey sample to the population of hospital CEOs in the province, as non-respondents may be biased in systematic but unknown ways from the sample. Accordingly the results will be confined to the sample. 7.4 Hospital characteristics The reported bed counts in the survey sample ranged from 18 to 1265, with a median of 128 beds, and a median 1999/2000 operating budget of approximately 29 million dollars. Respondents were mainly from smaller hospitals. Equal proportions of respondents (44.2%, n=38) reported that their hospitals were organized along departmental and program lines, respectively, with 12% (n=10) using a combination of departments and programs. Slightly less than one third of respondents reported that their hospitals dedicated staff to planning and research on operations for strategic planning and resource-allocation. In the sample, models of hospital governance showed considerable variation in number of board members and meetings, but high consistency in openness. The median number of board members was 18, with the number ranging from 7 to 62 members. The median number of board meetings annually was 10, with a range from 4 to 20. The majority of respondents (84%, n=72) indicated that their hospital’s board meetings were open to the public, while 85% (n=73) indicated that board meetings were open to media. Most hospitals in the sample had undergone considerable changes in management. A majority of respondents (59%, n=51) reported that their hospitals had changed CEO during the previous five years, while 86% reported that there had been a change in management structure during the previous five years. Respondents were asked to indicate whether their hospitals were currently experiencing difficulties in each of several aspects of their operation. Fully 65% or more of respondents indicated that their hospital was experiencing significant difficulties, specifically:

• Operating budget, • Accommodation funding, • Clinical equipment funding, and • Recruitment of doctors, nurses, and other staff.

41

The only aspect of operation with which a majority of respondents did not report difficulties was accumulated debt. Table 2 shows the distribution of responses for each aspect of operation: Table 2: Proportion of respondents reporting significant difficulties at their

hospital

Significant Difficulty Aspect of Operation

Yes No

1. Accumulated debt 29% 71%

2. Operating budget 71% 29%

3. Accommodation funding 72% 28%

4. Clinical equipment funding 66% 34%

5. Recruitment of MDs 76% 24%

6. Recruitment of nurses 78% 22%

7. Recruitment of other staff 75% 25% The frequency of management changes and operational difficulties reported in the survey is consistent with the pattern of environmental complexity and uncertainty, and of operating difficulties, that respondents described in the key informant interviews.

7.5 Determinants, pressures and priorities Respondent CEOs were asked to rate the extent to which each factor (from a list of provided factors) constrained or facilitated resource-allocation in their hospital. The list of 28 factors was drawn from the key informant interviews, and respondents rated the effect of each on a five-point scale ranging from “highly constrains” to “highly facilitates” resource-allocation. Mean ratings and standard deviations for each of the 28 factors are shown in Table 3. Inspection of Table 3 reveals that the factors that respondents rated as most constraining resource-allocation were: (a) MOHLTC base funding, (b) negotiated salaries and wages, and (c) physical plant. Respondents rated the following factors as moderately constraining: (a) capital equipment, (b) staff recruitment, (c) access to long-term care, and (d) medical staff.

CEOs saw only a few factors as facilitating resource-allocation, with mean ratings higher than the scale midpoint. These facilitating factors were: (a) service provider partnerships and, alliances, (b) affiliation with a medical school, and (c) quality of care. Overall, more factors were seen as constraints than facilitators on resource-allocation. These findings are consistent with the description from the key informant interviews of determinants as having mostly constraining effects on resource-allocation.

42

Table 3: Ratings of degree to which factors constrain or facilitate resource

allocation

Factor Mean Rating

Standard Deviation

1. MOHLTC funding 1.64 1.02

2. MOHLTC protected program funding 2.45 0.88

3. MOHLTC priority program funding 2.89 1.05

4. HSRC directives 2.64 1.00

5. Funding for HSRC directives 2.49 1.15

6. Hospital network(s) 3.00 0.72

7. Service provider partnerships/alliance 3.47 0.71

8. Community Need 3.13 1.17

9. Geographical location 2.67 0.99

10. Service providers/facilities in catchment area 2.92 0.91

11. Competition for market share 2.92 0.66

12. Access to long-term care resources 2.30 1.02

13. Special interest groups 2.95 0.49

14. Medical school application 3.14 0.76

15. Staff recruitment 2.34 0.92

16. Negotiated salary and wages 1.96 0.87

17. Medical staff 2.44 1.06

18. Capital resources 2.53 1.25

19. Physical plant 2.01 0.98

20. Capital equipment 2.26 0.95

21. Program change 2.98 0.86

22. Support services 2.92 0.68

23. Quality of care 3.38 0.98

24. New technology 2.76 1.10

25. Research activity 3.04 0.59

43

Factor Mean Rating

Standard Deviation

26. Teaching activities 3.10 0.59

27. Fundraising capabilities 3.12 1.13

28. Wealth of community 2.94 1.11

7.6 Decision-making processes Respondents were asked to rate the usefulness of various formal planning methods, data and information tools for resource-allocation decision-making. They were also asked to rate the usefulness of various formal procedures and practices to guide patient care decisions. Ratings on a five-point scale were provided for each type of procedure ranging from “Very Low” to “Very High.” An option for “Does not apply to this hospital” was also provided. The formal planning procedures that respondents rated for usefulness were:

(a) Mission statement; (b) Cyclical strategic planning; (c) Organized priority-setting for resource-allocation; (d) Policy manual; and (e) Clinical decision guidelines.

Respondents’ ratings of the usefulness of formal planning procedures for resource-allocation decisions at their hospitals tended to be “Moderate” or “High” -- with the exception of policy manuals which were rated “Low” to “Very Low.” The median usefulness rating for each of the methods was “High” (4.0) except for policy manuals (“Low”, median=2.0) and clinical decision guidelines (“Moderate”, median=3.0). Table 4 displays the mean ratings and standard deviations of the ratings for each kind of formal planning method. Table 4: Mean ratings and standard deviations, usefulness to resource allocation decisions of formal planning methods

Planning Method Mean Rating Standard Deviation

1. Mission Statement 3.80 1.02

2. Cyclical strategic planning process 3.87 0.89

3. Organized priority setting process for resource allocation 3.89 0.88

4. Policy manual 1.91 0.86

5. Clinical decision guidelines 3.35 1.08

44

Analyses of variance were conducted on usefulness ratings as a function of hospital size. Hospitals were grouped into one of three categories, small (fewer than 100 beds), medium (100-400 beds), and large (more than 400 beds). Although mean ratings for the usefulness of policy manuals were low overall, the ratings were significantly decreased as hospital size increased. In other words, respondents at smaller hospitals gave higher usefulness ratings to policy manuals than respondents at medium and large hospitals (p < .05). Similarly, respondents at small and medium hospitals gave higher usefulness ratings to clinical decision guidelines than respondents at larger hospitals (p < .046). Turning to the usefulness of data and information tools for resource-allocation decisions, respondents gave high ratings to each of four kinds of tools: financial data; patient data; financial forecasting tools; and service-level forecasting tools. Mean ratings and standard deviations are presented in Table 5.

Table 5: Mean ratings and standard deviations, usefulness of data and information tools to resource allocation decisions.

Tool Mean Rating Standard Deviation

1. Financial data 4.33 .69

2. Patient data 4.03 .80

3. Financial forecasting tools 3.86 .91

4. Service – level forecasting tools 3.64 .80

45

Turning to formal procedures to guide decisions about patient care, respondents tended to give “Moderate” to “High” ratings of their usefulness. Table 6 displays the means and standard deviations of the ratings for each of these decision procedures for patient care. Table 6: Mean ratings and standard deviations, usefulness of decision

procedures for patient care

N Mean S.D. N/A

1. Concurrent patient audit 63 3.27 0.97 21

2. Retrospective patient audit 81 3.19 0.84 3

3. Severity index for emergency admission 67 3.61 1.00 15

4. Appropriateness review for non-elective admissions 62 3.15 0.96 21

5. Clinical pathways or care maps 76 3.61 1.06 8

6. Procedure for allocating OR time 60 3.55 1.62 22

7. Procedure for allocating ICU staff 53 3.45 0.99 27

8. Technology acquisition plan 71 3.63 0.83 13

9. Physician input analysis for granting priviledges 72 3.14 1.09 12

10 Drug utilization review 81 3.58 0.91 3

11 Opinions of staff ethicist 38 2.89 1.18 46

12 Opinions of national ombudsman 44 3.00 1.26 37 Overall, respondents indicated that formal planning methods, procedures and tools were moderately- to highly-useful for making resource-allocation decisions. At the same time, the low frequency of very high ratings for the usefulness of formal planning methods suggests that respondents did not completely rely on them in their decision-making. 7.7 Power and influence in resource-allocation decision-making The key informant interviews highlighted the diffuse nature of accountability in hospitals, inherent in a professional bureaucracy, created by the dual authority structure of administration and doctors. Further insight into the process of priority-setting and resource-allocation decision-making might be gained from an examination of power and authority relations in hospitals. Accordingly, the survey asked about the locus of authority for significant resource-allocation decisions, assessed the structure of hierarchical authority and the extent of external influences on resource-allocation priorities at respondents’ hospitals. From the interviews, it might be expected that doctors and clinical staff would be seen to have considerable influence on priority-setting, as would numerous sources external to the organization.

46

Respondents were asked to indicate who, among governors and senior management, made the majority of significant resource-allocation decisions. More than half of respondents (54%, n=45) identified their Boards, followed by 35% who identified management as making the majority of significant resource-allocation decisions. Clinical staff, Executive Committees and Board Committees was rarely identified as significant decision-makers (by 4%, 6% and 1% of respondents, respectively). The degree of perceived hierarchy (i.e. authority structure, where authority is power associated with position level) was assessed by asking respondents to rate the influence of each of the different major position levels on resource-allocation priorities at their hospital. This method was adapted from Tannenbaum’s (1968) technique for measuring hierarchical control in organizations (see also Katz and Kahn, 1978). The present study used the five-point rating scale developed by Tannenbaum, ranging from “Little or No Influence” to a “Very Great Deal of Influence.” The ratings of the influence of different position levels can be displayed in a ‘control graph,’ showing the perceived distribution of control in the organization, with mean influence rating on the vertical axis and position level in descending order on the horizontal axis. A control graph showing CEOs’ mean ratings of the influence of various position levels in their hospitals is presented in Chart 1. This control graph reveals a traditional hierarchical gradient of control, as evidenced by a negative slope, with higher-level positions exercising more power than lower-level positions. Importantly, CEOs perceive themselves as having the greatest influence on resource-allocation priorities, followed by the chief medical executive, the board, and the chief nursing executive at roughly the same level, and then the remaining position levels in descending order according to their location in the hierarchy.

Figure 1: Hospital CEO control chart

Hospital CEO Control Chart

00.5

11.5

22.5

33.5

44.5

5

Board

Senior

Man

agem

ent

Chief M

edica

l Exe

cutiv

e

Chief N

ursing

Execu

tive

Medica

l Staf

f

Nursing

Staff

Other C

linica

l Staf

f

Other N

on-C

linica

l Staf

f

Position Level

Mea

n In

fluen

ce R

atin

g

47

It is not possible to assess the extent to which group bias and error might have influenced the ratings reported by this sample of CEOs because other position levels (such as doctors, nurses, the chief medical executive, and chairman of the board) were not asked for ratings of their perceptions. Further study of the perceptions of hierarchical control by other position levels in hospitals might lead to increased understanding of the nature of power at the level of the organization. The key informant interviews identified a large number of external organizations or groups that may influence resource-allocation priorities in Ontario’s hospitals. Accordingly, respondents were asked to rate the influence of twelve kinds of organizations and groups on resource-allocation priorities at their hospitals using the Tannenbaum rating scale of influence. However, since there is no agreed-upon hierarchy among the different external organizations and groups, means and standard deviations for the influence ratings of external organizations and groups are presented in a table (see Table 7) and not as a control graph. The only organization or group that respondents rated as having considerable influence -- from the list provided -- was the Ministry of Health and Long Term Care. Strikingly, respondents did not perceive any of the other organizations or groups as exercising substantial influence on resource-allocation priorities at their hospitals. Table 7: Mean ratings and standard deviations, influence of external

organizations and groups on resource allocation priorities

Organization or Group Mean Rating Standard Deviation

1. MOHLTC 3.67 1.12

2. Consultants 1.78 0.71

3. Community MDs 2.29 1.07

4. Professional colleges 1.54 0.61

5. Municipal government 1.45 0.65

6. Community care 2.08 0.76

7. Member of Provincial Legislature 1.51 0.84

8. General public 2.25 0.86

9. Media 1.53 0.70

10 Hospital networks 2.25 0.91

11 OHA 1.71 0.70

12 DHC 1.76 .68 Mean ratings and standard deviations, influence of external organizations and groups on resource-allocation priorities

48

7.8 Accountability for Reasonableness Respondents were asked to rate the priority setting decision-making at their hospitals using the ethical framework of accountability for reasonableness (Daniels & Sabin, 1997). Responses to these items were used to identify sites for conducting case studies. This is reported in detail in the following section on the case studies. 7.9 Goals of priority setting Respondents were asked, “What do you see as the goal(s) of priority setting in your hospital?” A modified content analysis was used to identify the most prevalent goals. The most cited goal (n=39) was, “To use resources to provide services to meet patient needs” (used in conjunction with ‘effectively’ and ‘efficiently’). The second most cited goal (n=17) was, “to align (or focus) resources to organization goals / strategic plan”. The third most cited goal (n=10) was, “to balance patient need, quality care and effective resource use.” Other goals, cited six times or fewer, included: balance budget; allocate resources for maximum benefit; ensure staff satisfaction and just treatment; find money for new clinic, buildings, staff, equipment; meet future requirements; improve health status of community; develop networks; the equitable distribution of resources; performance evaluation; improve cost-effective patient care; and ‘don’t understand’.

7.10 Rating Priority Setting Decision Making Respondents were asked to rate the priority setting at their own hospital in two ways. First, they were asked to provide two overall ratings of (a) how well they met their priority setting goals, and (b) how fair was their hospital’s priority setting. Second, they were asked to rate how well each of the four conditions of accountability for reasonableness were met at their hospital. Only two hospitals rated themselves a ‘5’ on all of these questions -- these were the two hospitals chosen for case studies.

7.11 Summary of findings Results for the ratings are presented in Table 8 (mean rating and standard deviation). Figure 2 displays the frequency distributions of the responses to each of these questions.

49

Table 8: Mean rating and standard deviation

Question Mean Rating

Standard Deviation

1. How well does your hospital meets its priority setting goals(s)? 3.86 0.86

2. Overall, how fair is priority setting at your hospital? 3.71 0.85

3. How well is the publicity condition met at your hospital? 3.59 0.97

4. How well is the relevance condition met at your hospital? 3.92 0.83

5. How well is the appeals condition met at your hospital? 3.48 1.01

6. How well is the enforcement condition met at your hospital? 3.23 1.06

Figure 2: Distribution of Ratings

0

10

20

30

40

50

Q1 Q2 Q3 Q4 Q5 Q6

Question

Freq

uenc

y

Rating=5Rating=4Rating=3Rating=2Rating=1

7.12 Conclusions The considerable influence of doctors on resource-allocation decisions seen in the interviews was not evident in the survey. However, the overall pattern of survey results corroborates the interview findings that CEOs are in charge of key resource-allocation decisions.

50

8.0 CASE STUDY; ACCOUNTABILITY FOR REASONABLENESS Priority setting (also known as rationing or resource allocation) can be defined as the distribution of resources among competing programs or people (McKneally, 1997). It occurs at all levels of the Canadian health care system, including the federal and provincial governments, regional health authorities (RHAs), provincial pharmaceutical benefit-management organizations, provincial disease networks, hospitals, and clinical programs. This project focuses on hospitals. The two interrelated objectives of this project are to:

1. Describe priority setting in a selected Ontario hospital -- we will describe the process of decision making by senior management related to each hospital’s annual budget cycle (i.e. what decision makers ‘do’).

2. Evaluate this description against a conceptual model for legitimate and fair priority setting -- we will compare the description of what decision makers ‘do’ with the leading ethical framework for legitimate and fair priority setting, “accountability for reasonableness”, which describes what they ‘should do’, and identify the gaps.

This study combines two relevant methods: case study research to describe priority setting and interdisciplinary research to evaluate the description against an ethical framework -- accountability for reasonableness. It is important because there is only one limited description of priority setting in hospitals, and priority setting in hospitals has not been evaluated using accountability for reasonableness. In this section, we will: 1) briefly describe the background literature relevant to priority setting in hospitals, and identify its gaps; 2) describe the ethical framework used in this study, accountability for reasonableness, and justify its use; 3) describe the methods used in this case study; 4) describe the study's key results; and, finally, provide a discussion of the results as they relate to the overall goals of this report. 8.1 Priority Setting in Hospitals Canadian hospitals account for one-third of health care expenditures. In1998, $27.6B of the $83.9B spent on Canadian health care was spent on hospitals, and over one-third of hospital expenditures, $10.1B, occurred in Ontario (http://www.cihi.ca/facts/nhex/hexdata.shtml). Moreover, all Canadian hospitals are stressed from several years of financial constraints, mergers and closures, as well as changing demand patterns resulting in, for example, surgical waiting lists, overcrowded emergency rooms, and an inadequate supply of resources for discharged patients. In particular, Ontario hospitals boards, which have responsibility under the "Public Hospitals Act" for stewardship of the organization and the funds received in transfer payments from the provincial government, struggle to fulfill their missions, provide quality care, and maintain the 'bottom line' (Macdonald, 1999). Due to both the prominence of hospitals within the health care system (e.g. over one-third of spending) and the significant difficulties they face (e.g. waiting lists), improving priority setting within hospitals is a key ethical issue for the Canadian health care system.

51

Although priority setting occurs at all levels of the health system, most priority setting research has focused on the macro (health system) or micro (bedside) policy-making levels. However, much of the priority setting in a health system occurs at the so-called ‘meso’ level of policy making, which includes RHAs and hospitals -- as mentioned, hospitals alone account for one-third of Canadian health spending. Although there are studies of priority setting in RHAs, (Williams, 2000, Yeo, 1998, Williams, 1996, Hurley, 1995, Bear, 1998, Ham, 1993, Hope, 1998), little is known about priority setting in hospitals. Singer and Mapa have described the factors relevant to hospital priority setting, including its mission, (Singer, 1998) and Kovac has described how government funding cuts in Australia have cause hospitals to ration their services (Kovac, 1998). However, these were commentaries and did not involve gathering data about actual practices. There are only a few data-based studies relevant to hospital priority setting. Taylor et al (1998). compared back and neck hospitalizations between Ontario and Washington State and found that hospitals in the two jurisdictions had different utilization and admission rates. Blundell and Windmeijer (2000) analyzed waiting times to develop a model of demand for acute hospital services in the U.K. Eland et al. (1998) examined 130 Dutch hospitals and found that access to taxoids varied greatly between hospitals. Alexander et al. analyzed the relationship between physician and hospital resources on hospital use in the U.S. (Alexander, 1999). However, none of these studies examined actual hospital priority setting. To our knowledge, only one study has examined actual priority setting in hospitals: Deber et al. described technology acquisition in Canadian hospitals (Deber et al 1994). However, this study was limited to a focus on new technologies, not on the entire range of hospital priority setting decisions. Summary: Although hospitals are significant foci for priority setting in the Canadian health system, there is only one limited study of actual priority setting in hospitals. The purpose of this case study is to describe and evaluate actual priority setting in an Ontario hospital. 8.2 The Ethical Framework: Accountability for Reasonableness Priority setting is primarily a problem of justice and, in the absence of consensus on substantive justice principles (what decisions should be made), the problem of priority setting becomes one of procedural justice (how decisions should be made) -- that is, a legitimate institution using fair priority setting processes. Legitimacy - moral authority over priority setting decisions - requires an institution to have a legal mandate to set priorities and follow a fair process. Thus, legitimacy and fairness are distinct but intimately related issues of justice (Daniels, forthcoming). The first step in understanding priority setting is to describe actual priority setting (i.e. what decision makers 'do'). But, because what decision makers 'do' may not be what they 'should do', it is necessary to go beyond description to evaluate the description of priority setting using an ethical framework. In this study, we chose accountability for reasonableness which is now the preferred conceptual model of leading priority setting researchers and decision makers internationally (Ham, 1999; Ham, 2000; Norheim, 2000; Martin, submitted).

52

Accountability for reasonableness is an ethical framework for legitimate and fair priority setting. It was developed by Daniels and Sabin the context of empirical case studies of priority setting in U.S. Health Maintenance Organizations and so is relevant to real-world decision making processes (Daniels, 1997). It is theoretically grounded in justice theories emphasizing democratic deliberation (Cohen, 1994, Rawls, 1993). No alternative ethical framework is both empirically and ethically grounded and focused on priority setting processes. According to accountability for reasonableness, an institution’s priority setting decisions may be considered fair if they satisfy four conditions: relevance, publicity, appeals, and enforcement, which are described in Table 9. Table 9: The four conditions of accountability for reasonableness Relevance Rationales for limit setting decisions must rest on reasons (including values,

principles and evidence) that fair-minded parties (managers, clinicians, patients, and affected others) can agree are relevant to meeting health care needs under resource constraints in the context.

Publicity Limit-setting decisions and their rationales must be publicly accessible. Appeals There is a mechanism for challenge and dispute resolution regarding limit-

setting decisions, including the opportunity for revising decisions in light of further evidence or arguments.

Enforcement There is either voluntary or public regulation of the process to ensure that the first three conditions are met.

8.3 Methods Our choice of research strategy is empirical research in bioethics that will combine two relevant methods: case study research to describe priority setting, and interdisciplinary research to evaluate the description against an ethical framework, of accountability for reasonableness. This strategy is appropriate because there is only one limited description priority setting in hospitals, priority setting in hospitals has not been evaluated using “accountability for reasonableness”.

8.4 Why we chose "describe-evaluate" as our research strategy Martin and Singer (2000) reviewed the current state of priority setting knowledge and proposed a new research approach in which they made four arguments. First, traditional discipline-specific approaches to priority setting (e.g. from philosophy, law, political science, medicine, and health economics) are insufficient because they are not grounded in actual experiences of priority setting in health care institutions, and the values that they contribute to priority setting conflict. Second, the current state of priority setting approaches can be conceptualized as a set of dialectal opposites: substantive criteria (e.g. efficiency, equity) can be contrasted with process criteria; and ‘simple solutions’ (Holm, 1998) (e.g. cost-effectiveness analysis) can be contrasted with ‘muddling through’, (Klein, 1998) an experimental and incremental policy making process. Third, what is now needed is a synthesis that integrates these dialectical opposites into a conceptual model incorporating both substantive and process criteria, and encompassing both ‘simple solutions’ and ‘muddling through’. Fourth, a research strategy to achieve this synthesis

53

will include a combination of empirical description using case studies and interdisciplinary evaluation using the best process-focused ethical framework. Ham and Coulter summarized international experience with health care priority setting, and concluded: (Ham, 2001) there is a need to strengthen institutional processes in which decisions are taken; priority setting processes must be transparent and accountable; there is a trend toward using clinical guidelines as a priority setting tool, but strong processes are needed for guidelines, just as for priority setting more generally; the politics of rationing may favour muddling through and the evasion of responsibility, but this is unsustainable in an environment of increasing public awareness about health care decision making; policy making regarding priority setting is an exercise in policy learning; and accountability for reasonableness is the leading ethical framework to evaluate priority setting processes in health care institutions. Accordingly, a strategy for improving priority setting in health care institutions entails improving priority setting processes, and accountability for reasonableness is the leading process-focused ethical framework to guide these improvements, and a strategy for improving priority setting at every level of every health system entails capturing and sharing the social policy learning that occurs in each context. Summary: Improving priority setting in health care institutions (e.g. hospitals) entails improving priority setting processes. A necessary first step involves describing their priority setting processes (using case study methods) and evaluating these processes (using interdisciplinary methods) using the leading conceptual model, accountability for reasonableness. This strategy is innovative because it can provide an integrated model for improvement that synthesizes substantive/process criteria and ‘simple solutions’/‘muddling through’. Finally, because priority setting in health care institutions is a problem at every level of every health system, an appropriate research strategy will also involve capturing the social policy learning that occurs so that it may be shared in other contexts.

8.5 Design This research used case study methods (Yin, 1994). A case study is an empirical method that seeks to uncover a phenomenon embedded in its real life context, particularly when the boundaries between phenomenon and context are blurred. Case study methods are appropriate because priority setting in hospitals is a complex, context-dependent, social process. Case studies are also advantageous in that they rely on multiple data sources, which allow for triangulation of data and the resulting thick description of the case.

8.6 Setting Orillia Soldier’s Memorial Hospital (OSMH) was selected for this case study based on their responses to the survey questions. In responding to the survey, hospitals rated how well they achieved their priority setting goals and how well they met the four conditions of accountability for reasonableness (see Table 9). Two of the 86 responding hospitals rated themselves as “5” on these five questions (scale 1-5 with 1 being not at all and 5 being does this extremely well). One of these two hospitals -- OSMH -- agreed to participate in the case study. The other hospital was

54

in the midst of a leadership change and was experiencing significant turmoil, and so did not agree to participate. OSMH is a 176-bed community hospital that services Orillia (pop. 30,000) and surrounding regions, with an estimated referral population of approximately 80,000. Operationally, OSMH functions under the Carver Governance Model, whereby the Board of Directors is responsible for overseeing the realization of Mission and Vision, and routine decision-making and planning is delegated to Executive Director (ED), Assistant Executive Directors (AEDs) and other senior staff. Communication regularly occurs between the Board and senior management in the form of monthly reports and participation of the ED and AEDs at monthly Board meetings. Senior management meets monthly at the Operations Committee (senior management operations and decision making committee) where all aspect of financial and organizational functioning are deliberated. The Operations Committee is the key priority setting forum in the senior management structure. A Capital Equipment Committee reports to the Operations Committee and is accountable for the dispersal of capital funds across competing demands. OSMH prides itself on operational efficiency and service to the community. In reviewing the Hospital Report 2001 prepared by the University of Toronto and the Canadian Institute for Health Information, OSMH ranked at average or somewhat above average on most indicators, with below average on one (access to diagnostic technology) and well above average for one (percent of total hours on direct patient care).

8.7 Data Collection The data sources for this study included documents and interviews with key informants. Documents reviewed included minutes of the Operations Committee for one year, Management Policies (including Management Limitations, Processes, CEO-Management Relationships and Organizational Ends or expected outcomes resulting form the hospital’s activities), Annual Reports (1998-99, 1999-2000), Operating Plan 2001-2001 and a document entitled “Creating a Healthier Future Together” that articulated a quality management program for OSMH. A total of six individual interviews were conducted with senior members of OSMH governance including the ED, three AEDs, the Chief of Medical Staff, and the Chair of the Board of Directors. An initial interview guide was developed based on the relevant literature and previous research (see Appendix 1, Interview Guide). 8.8 Data Analysis Documents, transcripts and fieldnotes were read, and memos were written to capture emerging concepts, ideas and relationships. Data were analyzed using open and axial coding techniques: (Strauss & Corbin, 1998; Strauss, 1994) open coding involved identifying portions of the narrative that related to key concepts; axial coding involved organizing concepts into broader categories providing an emerging description to be related to the priority setting and decision making. Rigour of the findings was enhanced by the use of multiple data sources (interviews and documents).

55

8.9 Results: Evaluation of OSMH Priority Setting using Accountability for Reasonableness We evaluated the priority setting process at OSMH according to the four conditions of accountability for reasonableness: relevance, publicity, appeals, and enforcement (see Table 1). Below, we will summarize key findings under each condition. 8.10 Relevance Key findings identified here included reasons, the role of participants, and the processes of decision-making.

8.11 Reasons Without exception, all interviewees referred repeatedly to the mission and vision of OSMH, and articulated service to the community as the key mission. Programs were expanded and partnerships established to service this priority. For example, the implementation and expansion of the dialysis program was intended to meet increased demands in the community and reduce the traveling patients were required to do to go to regional dialysis centres. Community was very broadly defined, as seen in OSMH setting up satellite programs in referring hospitals to meet the needs of that community. A major redevelopment of OSMH is planned over the next three to five years focused on the expansion and coordination of programs in response to the community. A recurrent theme in both the interviews and documents was balanced budgeting. Management policies specify that operational and capital expenditures may not exceed budget without justification and approval. Interviewees and documents reflected the trading off of costs and the use of innovations to save money in the long run as methods to achieve a balanced budget. Increased spending or “pushing the envelope” was possible only when other options or alternatives could not be identified. Increased spending was counterbalanced by the seeking of new revenue (primarily from the MoHLTC). Increased spending was accepted in priority patient circumstances. One interviewee described “I may lose my job over having a deficit, but I’ll never lose it because someone died on my watch”. “Pushing the envelope” with the MOHLTC was contentious at times, in one circumstance resulting in a standoff resolved only when a bank loan that facilitated increased spending for a patient care program was called due. The overall efficiency of the organization, coupled with its central role in health care provision in the region, likely contributed to the success of “pushing the envelope.” The role of data was central in reasoning about priority setting. Priority setting decisions were dependent on, and strengthened by, data. Clinical programs requesting more funds were instructed to bring data with their proposals. The ED provided much of his own data, both for presentations coming forward to the operations committee, and related to standards in the organization.

56

8.12 Participants The Operations Committee was described as the central decision making forum in the organization. However, interviewees described a very inclusive decentralized process of consultation and priority setting. Interviewees described open access to the Operations Committee, emphasizing the importance of proposals coming from frontline managers and staff. There appeared, however, to be a gate-keeping function on the part of the Assistant Executive Directors. Proposals from within programs were screened by the AEDs and “good” proposals would be brought forward to Operations. Proposals were deemed “good” if they met the priorities of the organization, and were adequately supported by data. There was, however, a phenomenon identified as the “squeaky wheel” whereby certain individuals or groups could directly access the ED. Participants were primarily members of senior management. Public input was primarily limited to a complaint process. Clinicians participated in priority setting by invitation only. While the broader community was clearly represented on the OSMH Board, the Carver Governance Model actually mitigates against board involvement in operation/budget decisions of the hospital. 8.13 Decision Making The management philosophy of OSMH is that decision making is a participative process, and that decisions should be made by those who are held accountable for the results of the decisions. Further, decisions must be compatible with the philosophy and objectives of the organization. Despite the inclusive process of consultation regarding priority setting, decision making occurred in a very centralized manner at the Operations Committee. Interestingly, despite the central role of the Operations Committee in priority setting, two of the six members described that they were not actively involved in the decision making, but rather that they would provide some input and then sit back and allow others to decide. These two AEDs commented that they took a passive role in decision making because they had neither a clinical or management background. One of the interviewees, who were relatively new to the organization, described difficulty in understanding the priority setting process and reasoning because it was an unstructured process with no formal written guidelines. Leadership style appeared highly influential in the process of priority setting decision making. The ED was seen as very accessible and hands on, integral to operations in the organization. Direct access to the ED could result in rapid decisions around spending.

57

8.14 Publicity Communication regarding priority setting occurred within the institution. The mission and vision of the hospital is widely disseminated throughout the organization. The minutes of the Operations Committee were available within the organization. The methods by which the minutes were disseminated were unclear, although the AEDs were accountable for distributing the minutes within their programs. Rationales for decisions were intermittently included in the minutes. Communication from the Capital Equipment Committee also occurred in a formal manner. The results of the committee’s deliberations were circulated widely within the organization. Rationales for decisions were not included with the decisions. The organization did not make its decisions and rationales accessible to the public.

8.15 Appeals The OSMH Operations Committee has a formal appeal process. However, the appropriate grounds for appeals are not specified. Appeals are presented either directly by the committee, or indirectly to the ED. It was not clear how often this occurred or how often decisions were reversed. It was unclear the effect of the different types of appeals (Operations Committee versus ED) on the final outcome of the decision. A formal appeals process existed for decisions of the Capital Equipment Committee. The decision list was circulated within the organization and members had a 30 day window in which to appeals the decision. Appeals of decisions made by the Capital Equipment Committee were heard at the Operations Committee.

8.16 Enforcement Condition The senior managers have made a concerted effort to base their decisions on principles -- aligned with the hospital’s mission and vision -- and accurate data. They have tried to be inclusive, allowing many others to participate in discussions about priority setting. The Operations Committee makes its minutes available throughout the organization. Proposals are encouraged to flow from the frontlines to senior management although the role of gate-keeping is unclear. Appeals are heard on priority setting decisions of the Operations Committee and Capital Equipment Committee. The Operations Committee is accountable, through the ED to the Board, to ensure that organizational goals are met. 8.17 Discussion In this case study, we described and evaluated the priority setting processes at OSMH. OSMH is an example of ‘good-practices’ in regard to priority setting, both by self-rating (see Survey Results above) and as evaluated in this study using the conditions of accountability for reasonableness. To summarize: They base their priority setting decisions on reasons (information and principles) considered relevant to the context by stakeholders (relevance condition); they make their decisions and reasons accessible throughout the organization (publicity condition); they have instituted an appeals process (appeals condition), and senior management have made an effort to ensure that these conditions are met (enforcement condition).

58

However, fairness (and legitimacy) is a relative goal, lying along a spectrum from unfair to very fair. Even if a particular institution appears to ‘do’ priority setting well (i.e. fairly), its processes may still be improved (i.e. made more fair). The key outcome is not ‘where’ the institution finishes in terms of a pre-determined standard, but whether it can improve. OSMH, as would be the case for any hospital, can improve their priority setting processes by doing more to meet the conditions of accountability for reasonableness. For example, they should: include patients, family members and community representatives on their Operations Committee thus enhancing the range of reasons for consideration; publicize their priority setting decisions and reasons outside the organization to patients and the public; specify the grounds for which appeals will be heard; and continue to study and evaluate their own practices to identify further opportunities for improvement.

8.18 Limitation The primary limitation of this research is its generalizability. Our results from a medium-sized urban/regional hospital may not be generalizable to other teaching hospitals, general hospitals (either urban or rural), or specialty hospitals (e.g. children’s hospitals). However, generalizability is seldom an all-or-none phenomenon and legitimacy and fairness are common goals for priority setting in every health care institution. It is likely that at least some of our lessons will be helpful in other contexts. 8.19 Conclusions OSMH graciously permitted us to study their priority setting processes and facilitated our collection of documents and interviews. Their courage in opening their administrative practices for scrutiny is commendable and admirable. Their senior management are leaders in health care administration and this study documents how that leadership translates into practice. This study at OSMH is a first step in what could be an on-going process of evaluation and improvement of priority setting at that institution. Changes could be made, based on the case study evaluation, and these changes should be studied in subsequent years to evaluate the effect they have on priority setting at the hospital. Since priority setting is a difficult problem for every hospital, what should others learn from this study? First, other hospitals could ‘see themselves’ in this description and evaluation and, therefore, learn from the findings. The findings of the OSMH evaluation may be similarly applied in other institutions. Second, this study shows the value of bringing a learning or continuous improvement perspective to priority setting in a hospital. This research platform presented here -- describe using case study methods and evaluate using accountability for reasonableness leading to recommendations for improvement -- can serve as a model for research-based reform to improve priority setting in hospitals. Third, because this study is the first of its kind, there is a need to do this in other hospitals and share the learning among all the hospitals. Therefore, The Change Foundation or the Ontario Hospital Association should develop a ‘clearing house’ of information about priority setting in hospitals where lessons could be shared and priority setting processes improved across the province.

59

9.0 CONCLUSIONS AND RECOMMENDATIONS Diversity and decentralization are strengths of Ontario’s hospital system. Diversity reflects the different circumstances, pressures and constraints characteristic of each setting, region, municipality and hospital management team. Decentralization is required to respond to the diversity of local needs. Decision-making and priority-setting for resource-allocation should be as much as is feasible matched to local pressures, constraints, needs and circumstances. The adoption of a single decision-making and priority-setting process on a province-wide basis may undermine the functioning of a system that -- by and large -- delivers good quality service in an efficient and accountable manner.

1. That The Change Foundation initiate research to determine what information would be most relevant to the decision-making for the priority-setting needs of hospital managers across Ontario and in what format such information would be most usable, and

a) That The Change Foundation identify and implement optimal means for disseminating such information to hospital managers in a frequent and timely manner.

2. That The Change Foundation initiate research to examine what priority-setting practices

work best for different types and sizes of hospital organizations in various regional settings, networks and alliances.

3. That The Change Foundation initiate research into how allocation decisions at the clinical-

level and program-patient level affect priority-setting at the hospital level; as well as how hospital-level priority-setting decisions impact upon the use of alternative resource choices.

4. That The Change Foundation initiate research to evaluate the influence of alliances and

networks on hospital-level priority-setting and resource-allocation.

5. That The Change Foundation initiate research to study the diversity of hospital boards of governance across Ontario by size, composition, membership rights/responsibilities and structure to identify preferred or optimal governance structures for hospitals by size, program mix and environmental complexity.

6. That The Change Foundation initiates a comparative evaluation of the effectiveness of

program and departmental management structures and processes for the delivery of clinical services.

7. That the Ontario Hospital Association create a face-to-face forum for continuous learning

and research transfer among hospital managers, decision makers and policy researchers concerning “what works best” among the specific needs and styles of Ontario’s different and distinct regions and hospital organizations.

60

8. That the Ontario Hospital Association in consultation with the College of Physicians and Surgeons of Ontario undertake to:

a) Solicit and support ongoing public participation in expressing the value bases of priority-setting and resource-allocation decisions;

b) Support and assist in the creation of guidelines for priority-selection in clinical decision-making at the patient service level through clarification of ethical issues; and

c) Promote the development of decision tools and guidelines based on explicit fairness and equity criteria, e.g. those of ‘accountability for reasonableness.’

9. That the Ontario Hospital Association collect the appropriate index information on a

regular basis and identify optimal means for disseminating such information to hospital managers in a frequent and timely manner.

10. That the Ontario Hospital Association and the Ontario Medical Association, in

consultation with the College of Physicians and Surgeons of Ontario, review and evaluate current practices, structures and processes involving the interaction between hospitals and community-based physicians and cooperate to develop, recommend and support protocols designed to improve communication between hospitals and the community of physicians in private practice.

11. That the Ministry of Health and Long-Term Care sanction and support hospitals, where

appropriate, to diversify their programs to respond to and address emerging issues in non-traditional ways.

12. That hospital managers:

a) Seek ways to allocate resources for the evaluation and improvement of internal information systems;

b) Support the development and use of information systems that produce comparable and compatible data among groups of hospitals;

c) Seek out, evaluate, and acquire resource utilization and forecasting software;

d) Seek out and adopt best practices for communication with community-based stakeholder groups and the MoHLTC.

61

APPENDIX 1

Case Study Interview Questions

1. Please describe priority setting at OSMH. (e.g. How does it work? Who is involved?)

2. In you view what is the goal of priority setting in your institution? Please explain.

3. Please describe your role in the process.

4. What principles or values are used to guide decision making at OSMH? Can you provide an example?

5. After decisions are made, to whom are the decisions made accessible? How are they

made accessible? (Can you provide an example?)

6. Is there an appeals mechanism for priority setting decisions? Please explain.

7. Is the process fair? Please explain.

8. How could this process be improved?

62

REFERENCES

Abelson, J., Lomas, J., Eyles, J., Birch, S., & Veenstra, G. (1995). Does the community want devolved authority? Results of deliberative polling in Ontario. Canadian Medical Association Journal, 153(4), 403-478.

Alexander, J.A., Lee, S.Y., Griffith, J.R., Mick, S.S., Lin, X., Banaszak-Holl, J. (1999).

Do market-level hospital and physician resources affect small are variation in hospital use? Medical Care Research & Review, 56, 94-117.

Anon. (1996). Emerging licensing issues in Canada and the US. International Nursing Review, 43(4), 11.

Atkinson, A. A., & McCrindell, J.Q. (1997). Strategic performance measurement in government. CMA Magazine, 20-23.

Bear, R., Frame, G., & Wetherill, S. (1998). The Last Critical Care Bed in Western

Canada. Healthcare Management Forum, 11(4).

Baker, R., Brooks, N., Anderson, G., Brown, A., MacKillop, I., Murray, M., & Pink, G. (1998-1999). Hospital Quarterly,2(2), 22-26.

Beach, L.R. (Ed.) (1996). Decision making in the workplace: A unified perspective. Mahwah, NJ: Lawrence Erlbaum.

Beach, L.R. (1998). Image theory: Theoretical and empirical foundations. Mahwah, NJ: Lawrence Erlbaum.

Blundell, R., & Windmeijer, F., (2000). Identifying demand for health resources using

waiting times information. Health Economics, 9, 465-74.

Bowling, A., Jacobson, B., & Southgate, L. (1993). Health service priorities: Explorations in consultation of the public and health professionals on priority setting in an inner London health district. Social Science & Medicine, 37(7), 851-857.

Canadian Institute of Health Information, National Information: National Health

Expenditures Database (NHEX). http//www.cihi.ca/facts/nhex/hexdata.shtmil.

Charles, C., & DeMaio, S. (1993). Lay participation in health care decision making: A conceptual framework. Journal of Health Politics, Policy and Law, 18(4), 881-904.

Chenoy, N. C., & Carlow, D. R. (1993). Changing perspectives on hospital governance. Healthcare Management Forum, 6(1), 4-10.

63

Coast, J. (2000). Explicit rationing, deprivation disutility and denial disutility: evidence from a qualitative study. In A. Coulter & C. Ham, The global challenge of health care rationing (pp.192-200). Buckingham, UK: Open University Press.

Coast, J., Donovan, J., & Frankel, S. (1996). Priority setting: The health care debate.

Chichester, UK: Wiley.

Cohen, M., March, J., & Olsen, J. (1972). A garbage can model of organizational choice. Administrative Science Quarterly, 17, 1-25.

Cohen J. (1994) "Pluralism and Proceduralism," Chicago-Kent Law Review, 69, 589-

618.

Conrath, D. W. (1967). Organizational decision making behaviour under varying conditions of uncertainty. Management Science, 13(8), 487-500.

Daft, R. L. (1998). Organizational theory and design, 6th Ed. Cincinnati, OH: South-Western.

Daniels, N., & Sabin, J. (1997). Limits to health care: Fair procedures, democratic deliberation, and the legitimacy problem for Insurers. Philosophy & Public Affairs, 4, 303-350.

Daniels, N., & Sabin, J.E., Setting limits fairly: Can we learn to share medical resources?

Oxford, UK: Oxford University Press, forthcoming. Deber, R., Wiktorowicz, M., Leatt, P., & Champagne, F., (1994) Technology acquisition

in Canadian hospitals: how is it done, and where is the information coming from? Healthcare Management Forum; 7(4), 18-27.

Dixon, J., & Welch, H. G. (1991). Priority setting: Lessons from Oregon. The Lancet, 337, 891-894.

Donovan, J., & Coast, J. (1994). Public preferences in priority setting - unresolved issues. In M. Malek (Ed.), Setting priorities in health care (pp 31-46). New York, NY: John Wiley & Sons.

Dyer, B.,(1996). The Oregon option: Early lessons from a performance partnership on building results-driven accountability. US Department of Health and Human Services. Revised March 3, 2002. http://aspe.os.dhhs.gov/progsys/oregon/lessons.htm.

Edgar, W. (2000). Rationing health care in New Zealand: How the public has a say. In A. Coulter, & C. Ham, The global challenge of health care rationing. Buckingham, UK: Open University Press, 175-191.

Eland, I.A., Otter, R., Krommendijk, R., & Stricker, B.H., (1998). Provision of taxoids in

1996: inequality of care [Dutch]. Nederlands Tijdschrift voor Geneeskunde, 142, 518-21.

64

Frankel, S., Coast, J., & Donovan, J. (1994). Health care requirements - A framework

for progress in priority setting. In M. Malek (Ed.), Setting priorities in health care (pp 103-114). NY: Wiley.

Freemantle, N., & Watt, I. (1994). Assisted conception techniques - On what basis do health technologies become routinely available when they have been assessed as effective? In M. Malek (Ed.), Setting priorities in health care (pp 217-228). NY: Wiley.

Accountability for priority setting in health care organizations. Unpublished views.

Goold, S. D. (1996). Allocating health care: cost-utility analysis, informed democratic decision making, or the veil of ignorance? Journal of Health Politics, Policy and Law, 21(1),69-98.

Gore, A. (1990). Oregon's bold mistake. Academic Medicine, 65,634-635. Ham, C., (1993). Priority setting in the NHS: reports from six districts. British Medical

Journal, 307, 435-8. Ham, C., (1999). Tragic choices in health care: lessons from the Child B case. British

Medical Journal, 319, 1258-61. Ham, C., & McIver, S., (2000). Contested decisions: priority setting in the NHS.

London, UK: King’s Fund Publishing.

Ham, C., & Coulter, A. (2000). The Global challenge of health care rationing. Philadelphia, PA: Open University Press.

Hardorn, D. C. (1991). The role of public values in setting health care priorities. Social Science & Medicine, 32(7), 773-781.

Holm, S., (1998). Goodbye to the simple solutions: the second phase of priority setting in health care. British Journal of Medicine, 317,1000-1002.

Holmstrom, S., (1997). Sweden and priority setting. Swedish American Health Care Foundation. http://www.swahc.org/priority.htm.

Hope, T., Hicks, N., Reynolds, D.J.M., Crisp, R., & Griffiths, S., (1998). Rationing and

the health authority. British Medical Journal, 317, 1067-9.

Hunter, D., (1993). Rationing: dilemmas and practices. Birmingham, UK: National Association of Health Authorities and Trusts

Hurley, J., Birch, S., Eyles, J., (1995). Geographically-decentralized planning and management in health care: some informational issues and their implications for efficiency. Social Science & Medicine, 41, 3-11.

65

Katz, D., & Kahn, R. (1978). The social psychology of organizations, 2nd Ed. NY:

Wiley.

Kitzhaber, J. A. (1993). Prioritising health services in an era of limits: The Oregon experience. British Medical Journal, 307, 373-377.

Klein, G. (1998). Sources of power: how people make decisions. Cambridge, MA: MIT Press.

Klein, R. (1993). Dimensions of rationing: who should do what? British Medical Journal, 307, 309-311.

Klein, R. (1998). Puzzling out priorities. British Medical Journal, 317, 959-960. Kovac, M. (1998). Rationing of hospital services in the Australian health system.

Croatian Medical Journal, 39, 339-45.

Lindblom, C. (1954). The science of ‘muddling through’. Public Administration Review, 29, 79-88.

Lomas, J. (1997). Reluctant rationers: public input to health care priorities. Journal of Health Service Resource Policy, 2(2), 103-111.

Loughlin, M. (1996). Rationing, barbarity and the economist's perspective. Health Care Analysis, 4, 146-156.

Macdonald, J.K., Shortt, S.E.D. (1999) An inventory and analysis of accountability practices in the Canadian health system. Kingston, ON: Queen’s Health Policy Research Unit.

McCandless, H., & Wright, D. (1993). Enhancing public accountability. Optimum, 110-118.

McKneally, M.F., Dickens, B.M., Meslin, E.M., & Singer, P.A. (1997) Bioethics for

clinicians: resource allocation. Canadian Medical Association Journal, 157, 163-167.

March, J.G. (1994). A primer on decision making: how decisions happen. NY: Free Press.

Martin, D. K., & Singer, P. A. (2000). Priority setting and health technology assessment: beyond evidence-based medicine and cost effectiveness analysis. In A. Coulter, & C. Ham, The global challenge of health care rationing, (pp.135-145). Bukingham, UK: Open University Press.

Martin, D.K., Singer, P.A., Priority Setting for Health Technologies in Canada.

(submitted).

66

Maynard, A. (1994). Prioritising health care - dreams and reality. In M. Malek (Ed.),

Setting priorities in health care (pp 1-18). New York, NY: Wiley. McKeown, K., Whitelaw, S., Hambleton, D., & Green, F. (1994). Setting priorities -

science, art or politics? In M. Malek (Ed.), Setting priorities in health care (pp 19-30). NY: Wiley.

McKneally, M. F., Dickens, B. M., Meslin, E. M., & Singer, P.A. (1997). Bioethics for clinicians: 13. Resource allocation. Canadian Medical Association Journal, 157(2), 163-167.

Mechanic, D. (1995). Dilemmas in rationing health care services: The case for implicit rationing. British Medical Journal, 310, 1655-1659.

Mintzberg, H., Raisinghami, D. & Theoret, A. (1976). The structure of unstructured decision processes. Administrative Science Quarterly, 21, 246-275.

Mooney, G. (1987). What does equity in health mean? World Health Statistics Quarterly, 40, 296-303.

Mooney, G., & Jan, S. (1997). Vertical equity: Weighting outcomes or establishing procedures? Health Policy, 39, 79-87.

Mooney, G. (1998). "Communitarian claims" as an ethical basis for allocating health care resources. Social Science & Medicine, 47(9), 1171-1180.

Mullen, P. M. (2000). Public involvement in health care priority setting: are the methods appropriate and valid? In A. Coulter, & C. Ham, The global challenge of health care rationing (pp.163-174). Buckingham, UK: Open University Press.

New, B. (1996). The rationing agenda in the NHS. British Journal of Medicine, 312, 1593-1601.

Nord, E., Richardson, J., Street, A., Kuhse, H. & Singer, P. (1995). Maximizing health benefits vs. egalitarianism: an Australian survey of health issues. Social Science & Medicine, 41(10), 1429-1437.

Norheim, O. (2000). Procedures for priority setting and mechanisms of appeal in the

Norwegian health care system.” Presentation at the 3rd International Conference on Priorities In Health. Amsterdam, The Netherlands.

Olsen, J. A. (1997). Theories of justice and their implications for priority setting in health care. Journal of Health Economics, 16, 625-639.

67

Perrow, C. (1986). Complex organizations: A critical essay, 2rd Ed. Glenview, IL: Scott Foresman.

Rawls, J. (1993). Political liberalism, New York: Columbia University Press.

Rayner, M. (1986). Evaluation in Canada. Evaluation Practice, 7(1), 45-56.

Sawyer, D. M., & Williams, J. R. (1995). Core and comprehensive health care services:

3. Ethical issues. Canadian Medical Association Journal, 152(9), 1409-1411.

Schafer, A. (1999). A wink and a nod: A conceptual map of responsibility and accountability in bureaucratic organizations. Canadian Public Administration, 42(1), 5-25.

Schatzman, L., & Strauss, A. (1993). Field Research. Englewood Cliffs, NJ:

Prentive-Hall.

Schickle, D. (1997). Public preferences for health care: Prioritization in the United Kingdom. Bioethics, 11(3 & 4), 277-290.

Schwartz, L. (1999). University of Glasgow, Department of General Practice Year 2 Rationing Plenary. Retrevied March 3, 2002, from http://www.gla.ac.uk/Acad/GeneralPractice/ration.htm. 1-8.

Scott, W.R.,(1998). Organizations: rational, natural, and open sytems. Upper Saddle

River, NJ: Prentice Hall.

Singer, P.A., & Mapa, J. (1998). Ethics of resource allocation: dimensions for healthcare executives. Hospital Quarterly, 1(4), 29-31. Strauss, A. (1987). Qualitative analysis for secim scientists. Cambridge, UK: Cambridge University Press.

Strauss, A., & Corbin, J., (1998). Basics of qualitative research: techniques and procedures of developing grounded theory. Thousand Oaks, CA: Sage Publications Inc.

Strauss, A., & Corbin, J., (1994). Grounded theory methodology: an overview. In: Denzin, N.K., Lincoln, Y.S., eds. Handbook of qualitative research. Thousand Oaks: Sage Publications, Inc.

Stronks, K., Strijbis, A-M. Wendte, J. F., & Gunning-Schepers, L. J. (1997). Who

should decide? Qualitative analysis of panel data from public, patients, healthcare professionals, and insurers on priorities in health care. British Journal of Medicine, 315, 92-96.

Tayor, V.M., Anderson, G.M., McNeney, B., Kiehr, P., Lavis, J.N., Deyo, R.A., Bombardier, C., Malter, A., & Axcell, T., (1998). Hospitalizations for back and neck problems: a comparison between the Province of Ontario and Washington State. Health Services Research, 33(4 Pt 1), 929-45.

68

Ubel, P.A., DeKay, M. L., Baron, J., & Asch, D. A. (1996). Cost-effectiveness analysis

in a setting of budget constraints: is it equitable? The New England Journal of Medicine, 334(18), 1174-1177.

Ubel, P. A., Loewenstein, G., Scanlon, D., & Kamlet, M. (1996). Individual utilities are inconsistent with rationing choices: a partial explanation of why Oregon's cost-effectiveness list failed. Medical Decision Making, 16(2), 108-116.

Van der Wilt, G. J. (1994). Cost-effectiveness analysis of health care services, and concepts of distributive justice. Health Care Analysis, 2, 296-305.

Varian, H. (1990). A solution to the problem of externalities and public goods when

agents are well informed. Ann Arbor, MI: Michigan Center for Research on Economic and Social Theory, Paper 90-01.

Williams, A. (1985). Economics of coronary artery bypass grafting. British Medical Journal, 291, 326-329.

Williams, J.R., & Yeo, M., (2000). “The Ethics of decentralizing health care priority

setting in Canada. In Coulter A, Ham C. (eds.) The Global Challenge of Health Care Rationing (pp. 123-132). Buckingham, UK: Open University Press.

Willliams, J.R., Yeo, M., & Hooper, W.(1996). Ethics for regional boards. Leadership in

Health Services, 5, 22-6. Yeo, M., Williams, J.R., & Hooper, W., (1998). Ethics and regional health boards. In L.

Groarke (ed.), The Ethics of the New Economy. Waterloo, ON: Wilfrid Laurier University Press. Yin, R.K., (1994). Case study research: design and methods. Thousand Oaks, CA: Sage

Publications, Inc..