Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An examination of the ethical decision-making processes used in decisions to fund, reduce or cease funding… Evoy, Brian 2008

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


24-ubc_2009_spring_evoy_brian.pdf [ 867.32kB ]
JSON: 24-1.0066877.json
JSON-LD: 24-1.0066877-ld.json
RDF/XML (Pretty): 24-1.0066877-rdf.xml
RDF/JSON: 24-1.0066877-rdf.json
Turtle: 24-1.0066877-turtle.txt
N-Triples: 24-1.0066877-rdf-ntriples.txt
Original Record: 24-1.0066877-source.json
Full Text

Full Text

AN EXAMINATION OF THE ETHICAL DECISION- MAKING PROCESSES USED IN DECISIONS TO FUND, REDUCE OR CEASE FUNDING TAILORED HEALTH SERVICES  by BRIAN EVOY  BSW, Carleton University, 1996 MSW, McGill University, 1997  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Interdisciplinary Studies)  THE UNIVERSITY OF BRITISH COLUMBIA  December 2008  © Brian Evoy, 2008  ii Abstract Health authority administrators were interviewed for their perspectives on what makes a good health care system; on tailored population-specific services as a way to address health inequities; and on how they perceive themselves to be making good funding decisions on the public’s behalf. The qualitative descriptive research dataset includes 24 hour-and-a-half long interviews with administrators from four BC health authorities, health region documents, memos, and field notes. Participants support the continuation of a public health care system and all participants acknowledge using tailored services as a route towards reducing health inequities.  However, these identified services have not been evaluated for their overall effectiveness. When it comes to decision-making, participants describe using a series of governance and bioethical principles that help them frame what and how issues can be considered. Decision situations are framed in a way that informs them whether they need to use formal or informal processes. In both cases participants collect information that allows others to understand that they have made wise decisions. The Recognition-Primed Decision Model accurately reflects the intuitive processes that participants describe using during informal decision-making and portions of formal decision-making. However, in relation to formal decision situations, there is less alignment with existing decision- analysis literature. Seven practice and future research recommendations are provided:  iii 1. Increase health authority participation in intersectoral partnerships that address non-medical determinants of health. 2. Develop new strategies for addressing health inequities. 3. Evaluate the efficacy of using tailored services beyond their ability to remove barriers to access. In addition, increase focus on testing new strategies for reducing the inequities gap. 4. Enhance existing decision-making processes by including the explicit review of decision tradeoffs, value weighting, and mechanisms for requesting revisions. 5. Focus future research on developing and evaluating the usefulness of formal decision-making tools in health authority structures and their relation to decision latitude. 6. Launch a longitudinal research study that examines how health authority expert decision-makers use judgmental heuristics and how they avoid the negative effects of bias. 7. Commission public dialogue on shifting the current illness-based system to one that is wellness based.  iv Table of Contents Abstract ................................................................................................................... ii List of Tables ........................................................................................................ vii List of Figures……………………………………………………………………viii List of Abbreviations ............................................................................................. ix Acknowledgements..................................................................................................x Dedication ............................................................................................................. xii 1 INTRODUCTION .............................................................................................1 1.1 Decision-Making Research.......................................................................5 1.2 Tailored Population-Specific Services......................................................7 1.3 Summary of Chapters .............................................................................12 2 METHODOLOGY ..........................................................................................19 2.1 Epistemological and Ontological Assumptions ......................................20 2.2 Knowledge, Representation, and Authority............................................20 2.3 Co-Constructing Knowledge and Interpretation.....................................22 2.4 Language.................................................................................................25 2.5 Narrative .................................................................................................26 2.6 Research Question ..................................................................................28 2.7 Specific Assumptions..............................................................................29 2.8 Research Objectives................................................................................31 2.9 Mission Questions...................................................................................32 2.10 Sampling and Data Collection ................................................................32 2.11 Recruitment, Data Collection Tools, Methodological Limitations.........36 2.12 Data Analysis Plan..................................................................................37 2.13 Documents ..............................................................................................41 2.14 Research-Related Ethical Considerations ...............................................43 2.15 Research Evaluation Criteria ..................................................................44 3 LITERATURE REVIEW ................................................................................48 3.1 Situating the Research Topic in Relation to the Literature.....................48 3.2 Governance Ethics ..................................................................................49 3.3 Decision Making Frameworks and Models ............................................56 3.4 Naturalistic Decision Models and Judgmental Heuristics ......................67 3.5 Public Engagement .................................................................................73  v 3.6 Analyzing Power.....................................................................................77 3.7 Conclusion ..............................................................................................80 4 PERSPECTIVES ON A GOOD HEALTH CARE SYSTEM.........................83 4.1 Introduction ............................................................................................85 4.2 Demographics …….................................................................................85 4.3 A Good Health Care System ..................................................................88 4.4 Benefits of health authority structure......................................................97 4.5 Medicare program and additional priorities..........................................100 4.5.1 Accountability and transparency.....................................................101 4.5.2 Primary health care and mental health care ....................................102 4.5.3 Eldercare and capacity to self advocacy .........................................103 4.6 Health care system renewal strategies ..................................................105 4.6.1 We need to tinker with the health care system ...............................108 4.6.2 We need a complete health care system overhaul ..........................110 4.7 Conclusion ............................................................................................114 5 PERSPECTIVES RELATED TO THE USE OF TAILORED       HEALTH CARE SERVICES...................................................................116 5.1 Introduction...........................................................................................116 5.2 How sub-population services get considered by Health Authorities ....119 5.2.1 Government orders..........................................................................120 5.2.2 Elected official influence ................................................................122 5.2.3 Strong community/advocacy lobby ................................................122 5.2.4 Health authority planning and management advocates ..................123 5.3 Strategies...............................................................................................127 5.3.1 Tailored health services is a good stopgap measure .......................128 5.3.2 Tailored health services helps resolve issues of health equity........130 5.3.3 Tailored health services is a poor use of limited resources ............132 5.4 Two significant considerations .............................................................136 5.4.1 Election cycles ................................................................................137 5.4.2 Funding cycle timing ......................................................................138 5.5 Predicting the likelihood of service selection and cuts.........................140 5.6 Complications .......................................................................................142 5.7 Making a difference in small ways: ......................................................146 5.8 Conclusion ............................................................................................149 6 DECISION-MAKING STRATEGIES ..........................................................153 6.1 Introduction.........................................................................................1553  vi 6.2 Framing the decision…………………………………………………..155 6.3 Decision tools........................................................................................170 6.4 Evaluating Decisions ............................................................................174 6.5 Conclusion ............................................................................................179 7 DISCUSSION................................................................................................183 7.1 Introduction...........................................................................................183 7.2 Good health care system .......................................................................183 7.3 Beliefs about tailored health care services............................................186 7.4 Decision-making practices....................................................................189 7.5 Public involvement ...............................................................................202 7.6 Power ....................................................................................................203 7.7 Limitations ............................................................................................210 7.8 Conclusion ............................................................................................214 8 CONCLUSION..............................................................................................219 8.1 Introduction...........................................................................................219 8.2 Summarized Reserarch Objectives .......................................................222 8.2.1 How participants make sense of their funding decision-making,       and how participants come to understand decisions are ethical...........222 8.2.2 How decisions are made; alignment with existing  decision-making frameworks; add to this area of knowledge. .......223 8.2.3 What participants think about tailored services ..............................224 8.3 Practice and research recommendations ...............................................225 8.3.1 Practice Recommendation #1 .........................................................225 8.3.2 Practice Recommendation #2 .........................................................227 8.3.3 Practice Recommendation #3 .........................................................227 8.3.4 Practice Recommendation #4 .........................................................228 8.4 Future Research Recommendations......................................................229 8.4.1 Research recommendation #1 .........................................................229 8.4.2 Research recommendation #2 .........................................................230 8.5 Conclusion ............................................................................................232 References............................................................................................................233 Appendix A: Individual Interview Protocol ........................................................243 Appendix B: Individual Consent Form................................................................246 Appendix C: Chapter Structure of Results Chapters .......................................25050  vii Lists of Tables Table 2.1: Summary of Interview Distribution......................................................34 Table 2.2: Summary of Funding Decision Case Examples ...................................34 Table 4.1: Select participant characteristics...........................................................86 Table 4.2: Employment Position Level Compared by Gender ..............................88 Table 4.3: Proposed health system changes...........................................................94 Table 6.1: Factors that influence decision-style ..................................................161 Table 6.2: Formal decision-making structural tools ............................................171   viii Lists of Figures Figure 3.1: A Decision-Analysis Process Flow Chart ...........................................59 Figure 3.2: Fundamental Objective Hierarchy Diagram........................................62 Figure 3.3: Basic Decision tree..............................................................................63 Figure 3.4: Means-Ends Objectives Network........................................................64 Figure 3.5: Levels of Engagement and Associated Engagement Mechanisms .....77 Figure 4.1: Chapter structure .................................................................................85 Figure 4.2: Elements ..............................................................................................88 Figure 4.3: Support for Medicare and additional priorities .................................100 Figure 4.4: Two strategies....................................................................................105 Figure 5.1: Chapter outline ..................................................................................118 Figure 5.2: Awareness of needs ...........................................................................119 Figure 5.3: Three strategies for change................................................................127 Figure 5.4: Two significant considerations..........................................................136 Figure 5.5: Factors that protect services from cuts ..............................................140 Figure 5.6: Complications....................................................................................142 Figure 5.7: Four administrator-lead acts of resistance.........................................146 Figure 6.1: Chapter outline ..................................................................................154 Figure 6.2: Framing the decision .........................................................................155 Figure 6.3: Decision tools ....................................................................................170 Figure 6.4: Evaluation..........................................................................................174 Figure 7.1: Decision-making influences ..............................................................190   ix List of Abbreviations Canada Health Act: CHA General Practitioner: GP Health Employers Association of British Columbia: HEABC Health Service Delivery Area: HSDA Integrated Health Network: IHN Information Technology:  IT Ministry of Health: MoH Non-Governmental Organization: NGO Non-Medical Determinants of Health: NMDH Programme Budgeting and Marginal Analysis: PBMA Recognition-primed decision model: RPDM Social Determinants of Health: SDH United Nations Development Programme: UNDP World Health Organization: WHO   x Acknowledgements I gratefully acknowledge the time and support that Dr. James Frankish, Dr. Michael McDonald, and Dr. Susan Cox provided me throughout my PhD program. They have provided me with the conditions for learning that interdisciplinary research makes possible. Thanks are also extended to Sarah Fielden who provided me countless hours of support throughout my degree. Most importantly I acknowledge Scott Elliott for forcing me to describe my entire research project in five words or less.  xi Dedication For Scott Elliott and Marilyn Evoy.  1 1 INTRODUCTION For the past three decades the Medicare program has been said to be a cornerstone of Canadian identity (Kaplan, 2004). As recently as 2002, Canadians have reconfirmed that the activities of the Medicare program, which are legislated through the Canada Health Act, should continue (Romanow, 2002). The Act includes five program criteria (public administration, comprehensiveness, universality, portability, and accessibility) that govern provincial Medicare program funding transfers (Parliament of Canada, 2004). In the main, this legislation has been successful in creating a consistent level of access to medically necessary services provided in hospitals and by physicians across the country regardless of an individual’s race, class, gender, ability to pay, or location of residence in Canada. Staying healthy and thriving, however, requires additional considerations well beyond acute focused illness prevention strategies. Over time, provincial governments have assumed a greater role in funding and/or delivering healthcare services, including a variety of programs and initiatives beyond the bundle of services and initiatives covered under the Canada Health Act. Examples of services that fall outside of Medicare include prescription drug coverage outside of the hospital setting, provision of community heath nurses, and home care services. In addition, governments, researchers, and advocacy groups have continued to draw significant attention towards addressing population health outcomes and narrowing identified health-related differences between and within sub-populations. This has been expressed most pointedly in epidemiological research highlighting the role of Non-Medical Determinants of  2 Health (NMDH) (Frankish et al., 2007; Kosteniuk & Dickinson, 2003; Lavis, 2002), through the establishment of the Population Health Approach (Health Canada, 1999), and position papers on health promotion and disease prevention (Health Canada, 1986). During the 1980s and 1990s we also see a series of initiatives that are focused on bottom-up community-based health promotion (e.g. school-based healthy heart programs that teach children about the importance of healthy eating and exercise), top-down policy-driven population health strategies (e.g. seatbelt laws) (Lavis, 2002), and a consistent recognition that intersectoral collaborations are imperative in addressing what some researchers have come to call “wicked problems”. Wicked problems are complex social issues such as poverty or unsafe housing that require the coordinated efforts of multiple sectors or government departments in order to make a significant difference (Bradford, 2005). As early as the 1980s Canadians saw the proliferation of provincial documents attempting to establish health system roadmaps that would address health system ills (Lomas, 1997) that included but were not limited to growing health system costs, emergency department overcrowding, and waitlists. The most consistent provincial solution was to devolve health care delivery systems to regional health authorities (ibid.). The degree of responsibility that was devolved and the evolution of the originally structured heath authorities have changed in British Columbia (BC) and have usually aligned with provincial election cycles.  3 In (BC), the 1991 Royal Commission on Health Care and Costs report “Closer to Home” proposed a decentralized system of health care delivery as a way to better health planning and managing and coordinating the continuum of health care provided to British Columbians.1 “New Directions for a Healthy B.C.” followed with five directions for implementing this reform process: • placing greater focus on all determinants of health; • establishing opportunities for public participation in health system decision-making; • increasing the role of volunteers in the health system; • establishing health regions to bring health system priority-setting, planning, and delivery to the local level; and • increasing attention to the effective management of the new system to ensure ethical and financial accountability. (Wharf Higgins, 1999) This is a tall order for health region Board members and administrators. There have been some empirical studies examining health authority administrator perspectives on their responsibility to support services and initiatives to address the Non Medical Determinants of Health (NMDH) (Frankish et al., 2007), community engagement strategies used to elicit public opinions (Abelson et al., 2004; Abelson et al., 2003; Frankish et al., 2002; Litva et al.,  1 The continuum of health care, often called “cradle to grave” care, includes the flow of services such as primary health care, long-term residential care, and tertiary care.  4 2002), and a decade-old study that examined Board member governance perspectives (Lomas, 1997; Lomas, Woods, & Veenstra, 1997). In addition, one Canadian study explicitly asked health region representatives how they include public values in decision-making processes (Menon et al., 2003). However, there are no known studies that examine how health authority administrators make sense of their own decision-making processes. Put another way, when health administrators are asked to reflect on their decision-making processes, how do they describe these experiences in a way that is logical? In efforts to reduce this identified gap, my first objective attends to describing the conditions and data inputs that health authority administrators perceive to be important in health authority resource allocation decision-making processes. In addition, I will provide an analysis of the ethical dimensions that interview participants ascribe to the decision-making process. Normative and justice ethics figure prominently throughout this dissertation and bear defining prior to turning our attention to decision-making research and tailored population- specific services. Tom Beauchamp and James Childress (2001) define normative ethics as “a form of inquiry that attempts to answer the question, ‘Which general moral norms for the guidance and evaluation of conduct should we accept and why?’” (ibid., p. 2). Normative ethics can be contrasted with descriptive ethics. Normative ethics make the case for what ought to be whereas descriptive ethics is more interested in “what factually or conceptually is the case” (ibid., p. 2). With normative ethics, theories are used to name and give reason for why the identified moral norms are acceptable.  5 Justice ethics includes the parcel of ethical theories, principles, and standards that argue for “the fair, equitable, and appropriate treatment in light of what is due or owed to persons” (Beauchamp & Childress 2001, p. 226). Having the ability to make claims on due benefits or burdens requires us to use standards of justice. Beauchamp and Childress (2001) state that an identified injustice has occurred when wrongful acts or omissions occur, resulting in a situation “that denies people benefits to which they have a right or distributes burdens unfairly” (p. 226). John Rawls (2003) is arguably one of the Twentieth Century’s most well-known philosophers who pens from a theory of justice and distributive justice position. Issues of equality, equity, fairness, and justice form the basis for addressing health inequities that are outlined below in the subsection on tailored population-specific services. 1.1  Decision-Making Research During the 1980s, the literature shows that decision researchers pursued at least two prominent streams of research: the continued development of models and frameworks to maximize the development and use of rational decision-making procedures (Keeney, 1992; Keeney & McDaniels, 1999; McDaniels, 2000; McDaniels, Gregory, & Fields, 1999), and a relatively new program of research addressing the role of heuristics (short cut thinking), intuition, and emotions as they relate to decision-making processes (Bazerman, 2002; Böhm & Brun, 2007; Gladwell, 2006; Klein, 1998; Todd & Gigerenzer, 2000). An important element of this second body of research was a recognition that rationality is bounded,  6 meaning that a person or group will never have all of the information needed to make a completely informed decision. Within a bounded rationality, reliance on certain heuristics (e.g. availability), described in further detail in chapter three, can be employed by individuals to produce either good time-sensitive decisions or decisions intended to promote an individual’s self interests. The role of intuitive decision-models is to highlight the processes that people go through in split-second situations to make “good enough” decisions. Both decision-analysis with its focus on thoroughness and attention to values and multiple stakeholder engagement, and intuitive decision-making with its recognition of time-limited contextual considerations and the role of tacit decision-making expertise, ought to hold strong appeal to health authority administrators and researchers. However, until very recently, Canadian health system administrators have not explicitly made use of decision- analysis, nor intuitive decision-making research and frameworks. More frequent have been the development and use of normative ethics priority-setting tools, and reliance on quality improvement cycles embedded with accreditation processes (Daniels, 2001; Daniels, 1985; Gibson, Martin, & Singer, 2005; Vancouver Coastal Health Richmond Health Services Ethics Committee, 2002). Recently some health regions in BC and Alberta have demonstrated a growing interest in the use of health economics-promoted Programme Budgeting and Marginal Analysis (PBMA) (Halma et al., 2004; Mitton & Donaldson, 2003; Mitton & Donaldson, 2002). While priority-setting research continues to grow, current literature, conferences, and  7 health authority activities fall short of addressing the tensions created between mainstream and tailored population-specific healthcare services competing for the same financial resources (Halma et al., 2004; Jacobs, Marmor, & Oberlander, 1999; Martin, Abelson, & Singer, 2002; Vancouver Coastal Health Richmond Health Services Ethics Committee, 2002). My second research objective is to examine participant descriptions of how they make decisions, explore how these descriptions align with existing normative and descriptive decision-making frameworks (ethical decision-making models, accountability for reasonableness, decision analysis, and naturalistic decision models), and add to this area of knowledge. 1.2 Tailored Population-Specific Services Today BC health authorities are charged with the responsibility of delivering the bulk of health services through hospitals and community and residential programs. They must also provide the Ministry of Health (MoH) annual performance measurement reports that detail their progress on closing the health gap among and between its constituents (Government of British Columbia, 2005). Human rights and justice ethics drive this imperative (Braveman 2006; Commission on Social Determinants of Health, 2007). As Paula Braveman (2006) argues: Human rights are that set of rights or entitlements that all people in the world have, regardless of who they are or where they live. When we encounter the term “human rights,” most of us think of civil and political rights such as freedoms of  8 assembly and speech and freedom from torture and cruel or arbitrary punishment. However, human rights also encompass economic, social, and cultural rights, such as the right to a decent standard of living, which in turn encompasses rights to adequate food, water, shelter, and clothing requisite for health, as well as the right to health itself. International human rights agreements also include the right to participation in one's society and the right to dignified as well as safe working conditions. (p. 183) In agreement with Macinko and Starfield (2002), I contend that deciding to initiate, increase, reduce, or eliminate funding toward initiatives and services intended to address health inequities is a value-laden political act, and not a mere technical procedure. The acknowledgement of existing differences in the health of a given population and the subsequent attempts by a society to reduce or eliminate those differences falls under the study of what has been variously termed health disparities, health inequalities, and health inequities. Researchers have made a range of attempts at distinguishing between these terms while others have chosen to collapse the distinctions. For instance, Braveman (2006), an American researcher, uses the term “health disparity” to describe inequities in health, whereas Macinko and Starfield (2002), who are also situated in the United States, forward the term “health equity”. Others have challenged the use of the term “disparities” because, they argue, it downplays the role of the inequitable distribution of wealth, privilege, and prestige in producing the conditions for the emergence of health differences (Commission on Social Determinants of Health, 2007). Nevertheless, even definitions of health disparity such as those posed by Braveman include an acknowledgment of inequitable conditions that  9 increase the likelihood that certain classes of health differences are unfair and unjust. In a similar distinction, Pedersen et al. (2007) suggest that a health inequality may become a health inequity once we observe a variation between population groups. Such a variation should be considered “unequal” or a “disparity”. When this inequality is unfair, avoidable, and/or unjust, then it becomes a health “inequity” (Braveman, 2006; Macinko & Starfield, 2002; Pedersen et al., 2007). Normative ethics provides the foundation for these assertions and thus technical or scientific “fact” is not enough to determine when an inequality is inequitable or what portion of the inequality is unfair or unjust (Pedersen et al., 2007). For the purposes of this thesis, I propose to pursue the use of health equity and inequity because of its explicit attention to the role of human rights and social justice and its fit within a Canadian context. The following definition of health equity comes from the World Health Organization (WHO) and is documented in the Commission on Social Determinants of Health (2007) who quotes the WHO Department of Equity, Poverty and Social Determinants of Health: “[T]he absence of unfair and avoidable or remediable differences in health among population groups defined socially, economically, demographically or geographically” (Commission on Social Determinants of Health, 2007, p. 7). The Commission adds to this definition by saying that “in essence, health inequities are health differences which are: socially produced; systematic in their distribution across the population; and unfair. Identifying a health difference as inequitable is not an objective description, but necessarily implies an appeal to ethical norms” (ibid., p. 7). In addition, they  10 suggest that “[a] model of community or civil society empowerment appropriate for action on health inequities cannot be separated from the responsibility of the state to guarantee a comprehensive set of rights and ensure the fair distribution of essential material and social goods among population groups” (ibid., p. 18). Attention to the moral dimensions of health inequity echoes Margaret Whitehead’s perspectives that have influenced many of the writers in this field. Whitehead (1990) argues that the cause of the inequity must be considered unfair when compared with the rest of society. She provides three necessary conditions when judging whether a health difference can be considered unfair and unjust: when health is being damaged because a person's lifestyle is severely restricted; where there is inadequate access to public services and essential health services; and where people are being exposed to unhealthy and stressful working and living conditions. When these conditions are avoidable, preventable, and/or removable, it is established that there are sufficient grounds to rectify the identified injustices. In addition, and regardless of the initial cause, people who become sick are more likely to have low income. The reduction in ability to access an adequate level of income is preventable through social policy options and therefore this situation is also considered unjust. This line of reasoning, however, does not go far enough to consider the social impact of fully attending to these injustices within a finite amount of resources. Choosing to fully attend to one social good may come at the expense of addressing another. As such, what has been established is that these injustices are prima facie wrong.  11 While health authorities do not explicitly state their political or ethical motivations, one consistent way that health authorities attempt to reduce the health inequities and the associated life expectancy gap is by delivering or funding health services or initiatives that are tailored to the needs of specific sub-populations. This strategy can be considered an effort to address “vertical equity” (Macinko & Starfield, 2002; Ruger, 2007). Tailored population- specific services are those created to meet the needs of populations that are not well served by mainstream services in hospitals and community settings. These services and initiatives attempt to remove cultural, language, and other barriers to access, and often provide the services in a format unlike associated mainstream services and initiatives. However, it is essential to remember that each time an administrator makes any funding decision, s/he2 is implicitly or explicitly deciding for or against another use for those financial resources. As such, a third objective of this dissertation is to examine what health authority administrators think about tailored population-specific services as a way to address the health inequities gap. For instance, do the participants explicitly consider the tradeoffs that they must make when recounting these decision-making processes? Finally, a fourth closely associated objective leads me to explore how these same administrators perceive themselves to be making “good”  2 To help preserve the anonymity of the participants quoted in this document, no gender pronouns can be used. Instead of using the cumbersome “he or she” and “his or her” every time a singular pronoun is required, or rewriting all sentences to use the gender-neutral plurals “they” and “their,” I have used the shorter “s/he” and “his/her”. I accept that this usage is ad hoc and not supported by many style manuals, but I have resorted to it here out of necessity.  12 decisions related to funding, reducing, or eliminating funding to community-based tailored health care services on behalf of the public. In sum, four objectives drive my research: gain a better understanding of how health authority administrators make sense of their own funding-related decision-making processes; look at participant descriptions of how decisions are made align with existing normative and descriptive decision-making frameworks and add to this area of knowledge; explore what the research participants think about tailored population-specific services as a way to address the health inequities gap; and examine how participants perceive themselves to be making “good” decisions related to funding, reducing, or eliminating funding to community-based tailored health care services on behalf of the public. 1.3 Summary of Chapters In the next chapter, I describe the qualitative descriptive analysis research methodology used in this dissertation. I argue that knowledge is context-dependent, shaped by multiple histories, context, and time. I suggest that through the sharing of intersubjective experiences, multiple truths come to coexist. From this I conclude that truth is established through the confirmation of valid knowledge that is co-constructed by members of a stake- holding community who use communal tests to gauge its value. The reader is encouraged to evaluate my research findings through the use of primary qualitative research evaluative criteria that include: credibility, authenticity, criticality, and integrity. These criteria are  13 supplemented by a series of secondary evaluative criteria that include: explicitness, vividness, creativity, thoroughness, congruence, and sensitivity. Combining narrative and qualitative descriptive analysis provides an excellent fit with my epistemological and ontological positions and my four research objectives. Using structural unit analysis I examine each individual participant interview transcript for salient storylines by exploring the narrative orientation through the use of why, what, where, when questions, exploring narrative turning points and how complications were resolved. I then temporally graph each story as a way to visually explore the narratives. In addition, I explore the variety of narrative structures offered across my participant interview dataset. Next, I order my interpretations of the entire dataset through the use of thematic coding. I believe that this exploration leads to a useful typology of how health administrators make sense of their decision-making processes and their perspectives related to the usefulness of tailored services as a vehicle for addressing health inequities across and within populations. In chapter three, I provide a review of relevant literature that will offer a greater understanding of my dissertation findings described in later chapters. This review will provide the reader with a foundation in a variety of substantive and procedural decision- making models, key governance-related considerations, and the role of public participation in decision-making processes. As well, I introduce basic concepts and theories related to “power” and map a line of inquiry into its examination. Power will be examined in more detail in chapter seven.  14 I introduce my research findings in chapters four-six. Separately, the chapters provide a detailed description of participant informed strategies for change that shape the myriad of participating health administrator perspectives on what makes a good health care system, the value of providing tailored health care services to marginalized groups, and descriptions of the substantive and procedural considerations that they use to make resource allocation decisions. Collectively, these chapters provide a broader understanding about what it means to work as an administrator in BC health regions at this point in time. In chapter four, I provide participant descriptions of what makes a good healthcare system. I show that the majority of perspectives shared during the interviews and through supplied supporting documents attend to the meso- or health authority level. This should be expected given that my research project was designed to elicit these responses, thus demonstrating how levels of interpretation combined with project framing erase or narrow potential readings (Alvesson & Sköldberg, 2000). I show that in general the participants state their support of the goals of the Canada Health Act. Research participants also explicitly promote a series of bioethical and governance-related principles and considerations that guide the tough decisions that they must consider on a regular basis. Next, I move the examination onto the discussion of equitable access to health care services. In chapter five, I introduce three interrelated strategies for change: (1) the use of tailored health services is a good stopgap measure; (2) the use of tailored health services helps resolve issues of health equity; and (3) the use of tailored health services are a poor use  15 of limited resources. I also describe a dilemma initially implied in chapter four, namely that while many of the participants promote the importance of pursuing issues of social justice, it cannot come at such a significant cost that the administrators are not being accountable for public dollars spent. Often one micro solution to this dilemma is to perform everyday acts of resistance, achieved through the use of existing budgets and related to forwarding issues of equity in small ways. Attention turns to decision-making strategies in chapter six. In this chapter it becomes clear that participants understand that their decision-making processes are value-laden. Three key principles are discussed: fairness, appropriateness, and collaboration. Through the examination of case examples provided by study participants, I suggest that there are six contextual factors that shape the likelihood that a decision process will be either formal or informal. These factors include: time, resource distribution, stakeholder position, proposed model, and the degree of risk associated with the issue at hand. Formal processes are more likely to be used during larger initiatives with longer timelines especially where the issues are politically contentious. These processes align well with traditional planning cycles and are often acknowledged through the production of terms of reference documents, meeting minutes, business cases, as well as proposal scoring activities used during funding selection processes. I argue that these formal processes are not as likely to rely on as many procedural considerations as traditional decision-analysis, such  16 as the usage of value weighting, but will often use various modeling processes known to decision-analysis. Similar to decision-analysis, the administrator will compare options before coming to a decision. Participants quickly assess what is needed in a given funding situation, and frame a decision context by employing informal processes. Time is a central consideration during both cutting and funding situations. Mental simulations similar to those forwarded by Gary Klein’s (1998) naturalistic decision model are used to assess whether the decision will work well. Reliance on intuitive decision-making processes figure prominently here. In these informal processes, the administrators describe their periodic use of props or rhetorical devices found in the formal processes, but these are not as pronounced. In addition, participants may compare options; simply go with the first “sensible” choice that presents; or select a direction where most stakeholders are supportive or silent. I also provide participant descriptions of how they know that their decisions were good. Participants evaluate their decisions by examining procedural, substantive, and intended outcome-based considerations. While participants have acknowledged that health authority decision-making processes have become more formal in recent years, they also provide a sobering caution that following a formal decision-making process does not guarantee that the results will be implemented. The most commonly cited reason for this is that stakeholders can use social action, media outlets, and elected officials to pressure the  17 health authority to alter their chosen direction. In chapter seven, I explore the interesting dynamic whereby participants state the importance of population health promotion but provide minimal evidence in their chosen case examples that NMDH are being attended to in any significant way. I also revisit the use of power within health authorities and decision-making practices. I confirm that participants are supportive of the Canada Health Act and support the social contract. In addition, participants demonstrate that they support several of the Romanow Commission report recommendations. I conclude that in the main two groups of participants are supportive of increase in formal decision-making practices in health authorities: those who are very experienced and those who are very inexperienced. Experienced administrators appreciate formal practices because it allows them to show that they have made procedurally fair decisions. Those with less experience desire a similar outcome, but are more likely to apply formal decision-making practices to situations where informal decision-making processes would have sufficed. I highlight that one experienced participant shows concern about how often formal decision-making processes occur within health authority structures. This participant demonstrates that in larger health authorities, senior administrators often forgo decision latitude in order to show more procedurally fair results.  I demonstrate how power flows through the preceding chapters in both coercive and productive ways. I also suggest that local and national policies shape future states of being  18 and their ultimate impact on health outcomes. I explore a policy’s epistemic ability to erase possibilities from ever being expressed (Kearney, 1994; Commission on Social Determinants of Health, 2007; Alvesson & Sköldberg, 2000; Lukes, 1974), and examine where the discourse of health regionalization and health outcomes are situated “in the play between powers at the tactical and strategic level, rather than simply trying to establish that it is an expression or an effect of a certain global strategy” (Alvesson & Sköldberg, 2000, p. 229). Throughout this chapter I evaluate my three research assumptions found in chapter two. I show where my assumptions are proven correct and I expand my understanding based on research findings. Finally, in chapter eight, I conclude by evaluating my intended research goals and objectives.  Afterwards, I conclude with a series of four practice and three research-related recommendations.  19 2 METHODOLOGY The social sciences have always found themselves in a situation of constant reorganization, characterized by a multiplicity of directions. It is not a state of crisis in a Kuhnian sense, that is, of a period with competing paradigms located between periods with normal science. The condition of social sciences has been termed “pre-paradigmatic,” if we remain in the Kuhnian terminology. The social sciences have always been in this state and as a result are neither relatively cumulative nor relatively stable. (Flyvbjerg, 2001, p. 30) Herbert Dreyfus argues that social science does not experience paradigm shifts but rather “style changes” (Flyvbjerg, 2001). Referencing Dreyfus, Bent Flyvbjerg (2001) states that social sciences go through periods where various constellations of power and waves of intellectual fashion dominate, and where a change from one period to another, which on the surface may resemble a paradigm shift, actually consists of the researchers within a given area abandoning a “dying” wave for a growing one, without there having occurred any collective accumulation of knowledge. (2001, p. 30) In this sense, we see fashion-like style changes. I situate myself as a researcher across what I believe to be three overlapping “style changes”: constructivist-interpretive; critical, and feminist-poststructuralism. Within these style changes my ontological, epistemological, and methodological beliefs guide my actions (Denzin & Lincoln, 2005). In this chapter I begin by making explicit my ontological and  20 epistemological assumptions and demonstrate their alignment with my research methodology. I choose a qualitative research tradition that aligns with my theoretical orientation and research goals. This ensures that my research questions, methods, and analytical processes align in a way that yields the intended results. I extend what Chenail (1997) calls the “plumb line,” that is, ensuring that the chosen research methods are able to produce the answers driven by the research question, to also include recognition of the philosophical and ideological underpinnings that I as a researcher bring to the work. Next, I describe my research design and analytical processes. I end the chapter with a discussion of key considerations related to evaluating the value of qualitative research. 2.1 Epistemological and Ontological Assumptions For the purposes of this thesis, ontology is defined as examining the nature of human beings and reality, and epistemology is defined as questioning “the relationship between the inquirer and the known” (Denzin & Lincoln, 2005, p. 22). Below, I characterize key elements of my research-based assumptions. These include perspectives on the role of interpretation and how it is represented through language and rhetorical tools to construct truth/knowledge claims. 2.2 Knowledge, Representation, and Authority Knowledge is context-dependent (Alvesson & Sköldberg, 2000). It is shaped by  21 multiple histories, context, and time (Clandinin & Connelly, 2000; Riessman, 1993). Given this complex constellation of considerations, I argue that for any account of the social world, multiple truths will exist. As a result it is not possible to mirror reality or claim research objectivity. Therefore I propose a modest claim: while we cannot mirror reality, we can share common axioms and intersubjective experiences that are situationally located rather than universal. Acknowledging multiple interpretation of what we observe and come to know (Finlay, 2002; Geanellos, 2000) must resonate with the community of people who share a common frame of reference. Truth comes from the confirmation of valid knowledge that is co-constructed by members of a group of stake-holding community members. Negotiation and dialogue moves the truth towards communal tests of validity.3 Within this common frame, some interpretations will come to hold more value than others given a host of contextual factors that include their novelty-value in calling into question established ways of thinking (Alvesson & Sköldberg, 2000; Flyvbjerg, 2001). At the level of the interview, the researcher and participant co-construct knowledge  3 Sheila Jasanoff (2005) argues that these collective tests to the value of knowledge take on characteristics that are unique to a given culture and country: “modern technoscientific cultures have developed tacit knowledge-ways through which they assess the rationality and robustness of claims that seek to order their lives; demonstrations or arguments that fail to meet these tests may be dismissed as illegitimate or irrational. These collective knowledge-ways constitute a culture's civic epistemology; they are distinctive, systematic, often institutionalized, and articulated through practice rather than in formal rules” (p. 255).  22 through the dialogical process (Charmaz, 2002; Gubrium & Holstein, 2003; Smythe & Murray, 2000). Given that an interview participant's subjectivity and experience is continually assembled and modified, it is not possible to judge his/her responses as objective. While some parts of their storied experiences will stay similar, other parts will shift as a result of new experiences that have alerted them to consider their experience in a new light. “Rather, the value of interview data lies both in their meanings and in how meanings are constructed” (Gubrium & Holstein, 2003, p. 33). In my research, I suggest that the truth- value of the interview interpretations have the ability to strengthen in at least three points: during the initial dialogue between the researcher and the participant where understandings are clarified; when initial research findings attributed to the participant’s quotes are shared with the participant; and when aggregate findings located across the dataset are shared with a broader community of readers. 2.3 Co-Constructing Knowledge and Interpretation The act of interviewing shapes the answers provided by the participants. The questions asked are constructed in a way that will produce results that will be useful in future ways and places (Briggs, 2003). This is the act of recontextualizing the interview discourse. Interview material must be decontextualized in order for the data to become reconstituted as  23 recontextualized results.4 Charles Briggs (2003) states that [t]he recontextualization process provides another angle on the complexity of interview data. A statement that emerges in an interview is tied explicitly to the question that precedes it and generally indirectly to previous questions and responses, the road range of texts, agendas, and contexts that shape the questions and interviewing practices, and the anticipated uses of the data ... As the work of Mikhail Bakhtin (1981) would suggest, this recontextualization process informs each word that is spoken, such that the different contexts, vocabularies, styles, subject positions, and the like are built into what is said and how it is uttered. Responses are like crossroads at which multiple paths converge, with signs pointing in all directions. The power of researchers thus lies not only in their control over what takes place in the interview itself but particularly in their ability to use that setting as a site that is geared toward creating a broad field for circulation of discourse. (p. 248) Similarly, Andrea Fontana (2003) suggests that meaning is shaped by contextual, historical, biographical, and institutional influences. She encourages us to understand the research interview as a site of co-constructing knowledge, which relies upon the “active interview” — meaning that the interview is a dialectical sense-making process between the interviewer and the participant. Both the research participant and the researcher influence the  4 This is not meant to imply that the text should be severed from its context history and temporality. That direction is forwarded by Geanellos (2000) following Paul Ricoeur’s writings, and on those points, in relation to interpretive research, I part company with the pair of them. I firmly believe that history, context, and time must be integral in the sense-making act of interpretive qualitative research in order to be of value.  24 shaping of the participant's story (ibid.). The interview becomes the venue through which the participant, through a dialogue with the interviewer, takes fragmented everyday experiences and turns them into coherent explanations. This work relies upon using a common vocabulary understood by both participants and through the selective framing of what gets highlighted and what is eliminated during the process of shaping the narrative (ibid.). Examples of research/participant co-constructing perspectives flow through the interviews included in this dissertation. They appear in situations where the interview protocol shapes the discussion, where the participant chooses to shift the topic towards something of interest to him/herself,5 and where I pick up on an association and pose a follow-up question. For instance, at the end of one interview, I share in more detail my interest in examining how participants describe their ethical decision-making processes. While the interview was originally winding down, my statement spurs the participant on to speak for an additional six pages regarding his/her discomfort or uncertainty about whether or not s/he has truly considered the ethical issues related to his/her decision beyond the principle “do no harm”. S/he says that models exist — such as the decision-tree flow sheet that s/he shares with me — but also states that no one in his/her organization applies the tool  5 The reader is reminded of my earlier footnote: To help preserve the anonymity of the participants quoted in this document, no gender pronouns can be used, so I have used “s/he” and “his/her”. I accept that this usage is ad hoc and not supported by many style manuals, but I have resorted to it here out of necessity.  25 in practice. 2.4 Language My interest in the study of expressed lived experience is guided by Ricoeur's writings described in Freeman (1997): “[Narrative] is the most fitting and appropriate language we could use to comprehend human lives in culture and in time” (Freeman, 1997, p. 175). Language is a necessary requirement of the interview process and storying of people’s experiences. While language is a medium that establishes meaning, it also constitutes truth(s) (Riessman, 1993). When people share their experiences, their language does not mirror an objective reality, but rather their stories are “constructed, creatively authored, rhetorical, replete with assumptions, and interpretive” (ibid., p. 5). Alvesson and Sköldberg (2000) comment on the relationship between the use of language and its ability to depict a reality: The relative ability of language to tell us something about our individual and social reality — once certain linguistic conventions have been culturally established and are regarded as working ‘tools’ — can thus be taken seriously. Its ability to conceal and tie down certain phenomena, giving them an unduly robust and unequivocal charter, must be continuously borne in mind. Language can be regarded as a blunt and equivocal instrument, which shapes and ‘perspectivizes’ rather than depicting or being totally dissociated from phenomena ‘out there’ … Respect for the distinctive character of language and its limited ability to reflect ‘reality’ is important. (pp. 244-5)  26 For Alvesson and Sköldberg, then, we have the ability to comment on and make sense of lived experiences, but we are not able to completely capture all facets of any given situation. In this section, I have suggested that knowledge is context-dependent, shaped by multiple histories, context, and time. Truth is established through the confirmation of valid knowledge co-constructed by members of a stake-holding community who use communal tests to gauge its value. Through the sharing of intersubjective experiences, multiple truths coexist. These positions shape my qualitative research approach to research. 2.5 Narrative Narrative approaches are especially well-suited to the task of exploring how decision-making unfolds because they situate the complex social and moral realities that decision-makers encounter in time and space. This temporal and spatial contextualization provides depth to significant relationships, weight to felt responsibilities, and texture to daily life. As such, narratives offer the listener, as well as the teller, a sense of biographical integration. Further, in drawing attention to the way that stories are embedded in decision-making, narrative approaches allow us to glimpse the powerful role of cultural expectations in shaping the way that decisions are constructed, experienced and represented. By this I mean that the stories we tell about decision-making shape and reflect broader cultural narratives about what it means to be responsible, exercise one’s agency, and make a “good” decision. (Cox, 2003, p. 272) One aim of my research is to represent and understand the health authority administrator decision-maker experiences. The object of study is the stories that participants  27 tell in relation to funding, reducing funding to, or cutting funding to health services and initiatives designed for a sub-population. I examined how the individuals I interviewed “impose order on the flow of experience to make sense of events and actions in their lives” (Riessman, 1993, p. 2). History, context, and temporality are essential considerations and were integral to my analytical process (Clandinin & Connelly, 2000; Freeman, 1997). Following Catherine Riessman (1993), I put forward that narratives are more than just storage devices; they structure perceptual experiences and organize memory. In this dissertation, I have used research participant narratives as a route to understanding how these perceptions are experienced and organized (ibid.). The terms “story” and “narrative” have different meanings. While narratives are verbal or written texts that provide an account of one or more events and actions that are chronologically linked, a story requires more (Czarniawska, 2004). In a story, the narrative must also have a plot. Barbara Czarniawska defines a plot as a point where equilibrium in the account is followed by disequilibrium and then followed by a new similar but different equilibrium (ibid.). In so doing, it provides a logical connection between episodes — how things are connected and the structure that makes sense of events. Czarniawska states that most stories can fall into one of four types: romance, tragedy, satire, or comedy (ibid.). Through the use of a two-step data analysis process I designed my research to include descriptive, explanatory, and evaluative elements. It is descriptive in the sense that I have  28 analyzed the structure of how stories unfold. Each interview contained multiple narratives and stories. Some were more developed while others were started but ended abruptly as discussions shifted directions during the course of the interview. In the following chapters I will advance my analysis by offering explanations about how people make sense of their experiences. I will also compare the stories across my data set to examine the variety of narrative structures that emerged and order these as “strategies for change”.6 As a result, I believe that it has lead to a useful typology of how health administrators make sense of decision-making and the usefulness of tailored services as a vehicle for addressing health differences and inequities across and within populations.7 Lastly, I examined narrative structure of the salient stories provided by interview participants, in both text and graphic representation. Peer reviewed literature has been consulted to compare it to my findings (Bottorff et al., 2000). 2.6 Research Question My overall research interest relates to how health authority administrators located in BC health authorities make sense of situations where they are charged with the responsibility  6 “Strategies for change” is described in more detail below. 7 I acknowledge Cox’s 2003 work on how typologies can be useful in understanding how individuals make sense of difficult decision-making processes.  29 to fund, decrease funding, or cut all funding to a tailored health service or initiative for a sub- population. This is an excellent location to examine the following four areas: 1. What do health care administrators say about the usefulness of tailored services as a strategy intended to close the health outcomes gap between and within populations? 2. What values do health care administrators express in relation to what makes a good health care system? 3. What do health care administrators consider makes a funding decision a “good” or “bad” decision? 4. What decision inputs do participating administrators claim to use in a decision- making process? These stories about decision-making processes are situated in the complex moral and social relations that are located in a particular time and space (Cox, 2003). They also provide a textured understanding of the cultural conditions that shape the construction, representation, and experience of the decision (ibid.). Similar to Susan Cox (2003), I was especially interested in what these conditions tell us about what it means to construct a “good” decision. As well, I wanted to know how the decision-making processes described by the research participants aligned with or diverged from existing normative and descriptive decision-making models and frameworks. 2.7  Specific Assumptions I have approached my study with three explicit assumptions that have shaped my  30 research. I have drawn these positions from my personal experience as a middle management health care administrator working in a large BC health region. My experiences include being a practicing social worker, a community developer, a leader conducting large scale public involvement processes to inform health authority policy development, and as a manager working collaboratively with family doctors to design more effective services for people who have chronic diseases. First, I posit that when it comes to funding, reducing funding, or cutting funding to targeted health services and initiatives, health care administrators draw on implicit ethical principles framed by the Canada Health Act (Health Canada, 1986), and implicitly mirror a social contract perspective (Mendelsohn, 2002), based on the belief that the collective has a moral responsibility to support individuals impacted by unequal social relations (race, class, gender, etc.). The Canada Health Act and the population health approach contain the egalitarian clam that all lives are valued. Value is also placed on people living longer and healthier lives. While the Canada Health Act does not explicitly extend beyond ensuring medically necessary health services are provided to all Canadians, Health Canada and the provincial Ministries of Health continue to publicly promote the population health approach which acknowledges the importance of a host of medical and non-medical determinants of health in addressing health inequities and keeping people healthier for a longer period of time (Health Canada, 1986).  31 Second, while ethical principles expressed in the Medicare program and the population health approach may guide the administrator to consider funding sub-population health services and initiatives, I propose that clinical efficacy and economic efficiency are the driving forces behind funding these services, with little attention paid to ethical and social value (Terry, 2004). Third, I assume that formal decision-making procedures are seldom used on the ground in health regions unless it is a large-scale health service redesign. 2.8 Research Objectives My research objectives are to: • understand how diverse participants’ perceptions are experienced and organized (Riessman, 1993) and to “interpret the underlying, implicit meanings behind what they say” (Smythe & Murray, 2000, p. 324); • identify the criteria that health region administrators use in health authority resource allocation decision-making processes; • describe the ethical dimensions that interview participants ascribe to the decision- making process; and • examine how participant descriptions of how decisions are made align with existing normative and descriptive decision-making frameworks (ethical decision-making models, accountability for reasonableness, decision analysis, and naturalistic decision models) and add to this area of knowledge especially within a Canadian health authority context.  32 2.9 Mission Questions Ronald Chenail (1997) uses the term ‘mission question’ to refer to the actual questions that an individual wants to research. The following mission questions are the things that I wanted to know about my area of curiosity (research question). These are somewhat different than my actual interview protocol but the essence of these questions are contained therein. 1. How do interview participants describe the characteristics of a “good” health care system? 2. What do interview participants take to be the values involved in “good” decision- making? 3. How do health care administrators perceive themselves to be making funding decisions? 4. What inputs do interview participants see as important in decisions related to funding, reducing, or cutting funding to a tailored service?8 2.10 Sampling and Data Collection I interviewed 24 health care administrators from four health regions from across BC for one to one and a half hours each resulting in over 800 pages of verbatim transcript.9 I  8 See Appendix A for the individual interview protocol. 9 A fifth health authority refused to participate in the study.  33 selected the interview participants using the following criteria: they were required to be health authority administrators currently working in a BC health region10 and they must have been in a position to significantly influence11 the decision to fund, reduce, or cut funding to a tailored community-based health service or initiative12; they could identify a situation where they had to influence a funding decision related to a tailored health service or initiative; and the situation they described must occur in an urban or a rural setting. (See Table 2.1 for a summary of these selection criteria.) In each health region a minimum of three participants were interviewed about the establishment of a tailored community-based health service or initiative and a minimum three participants spoke to the full or partial cutting of tailored community-based health service or initiative funding. Each health region-specific set of  10 In BC there are six health regions: five serve a geographically designated area and one services the entire province. For the purposes of this study it was unlikely that the provincial-based health region would provide additional information that the geographically-based regions offered. Therefore I decided not to include the Provincial Health Services Authority. In addition, and as previously stated, one geographically designated health authority refused to participate in the study. Nevertheless, the participating health regions represent a significant amount of the population of BC residents. To conceal the identity of the region that chose not to participate, I have not provided a specific percentage of residents that the participating regions represent. 11 Often, several people make a final decision on a funding issue in a health region. Therefore I sought out individuals who described themselves as having an instrumental role in the funding decision that they recount. 12 Community-based health services are any services that are located outside of the acute health service system. They can be located in Community Health Centres, in non-profit society agencies, and/or delivered directly in the client’s home.  34 interviews was required to describe at least one urban and one rural experience. (Refer to Table 2.2 for a visual representation of the number of actual case examples.) Follow-up emails were used to clarify interview information and/or collect missing information. Table 2.1: Summary of Interview Distribution 1-1 Interview Categories 6 Per Region Urban 1 minimum per region Rural 1 minimum per region Decision to Fund a Tailored Service 3 per region Decision to Reduce Funding or Cut Funding to a Tailored Service 3 per region Total: 24 Table 2.2: Summary of Funding Decision Case Examples13 Urban Rural Fund Reduce/Cut 16 10 15 15 I informed potential interview participants via email and a subsequent phone call that no one, including any referring agent, would be informed about their participation or refusal  13 Some participants provided examples that aligned with more than one category. Therefore the total number of responses is higher than the actual number of interviewed participants.  35 to participate in my study. I also offered all potential participants several opportunities to restate their interest to participate in the research. These included: • an option on the first email to participate; • during the phone call where I screened the participant against the study inclusion criteria and provided them with additional information about the study; and • just prior to commencing the actual interview where they were reminded about the purpose of the study and asked to sign a consent form. At any point in this recruiting process the individual had the option to withdraw his/her interest in participating. However, once the participant actually attended the interview, his/her transcript data was used in the research analysis. Lastly, the study dataset was completed through the inclusion of personal field notes, memoranda, and relevant documents. These included the following sources: emails, reports, sharable internal health authority documents, and health authority website material. I asked the selected research participants for these documents. The interview protocol was piloted during the first three interviews. With the exception of one question,14 the interview protocol worked well and only minor changes  14 The question that was eliminated asked participants to first describe the most important data inputs that they believed were necessary for their decision and then to consider what would happen if that input was no longer available. In the situation where the participant was describing a funding decision, they consistently stated that they would not make the decision until they had that data input. In the cases where the participant  36 were made. 2.11 Recruitment, Data Collection Tools, and Methodological Limitations Limitations related to recruitment and data collection tools include access to participants, power dynamics, and cultural norms expressed throughout the interviews. Recruiting health region administrators required successfully negotiating agreements with each quasi-governmental health authorities prior to interviewing their administrators and collecting health authority documents that describe past decision-making processes. Without this approval, I would not have had access to research participants and this would have severely restricted my ability to generate insights relating to the chosen topic. To secure this agreement, I provided the health authority contacts with a clear and concise description of the research and discussed with them any concerns related to their employees participating in the study. My research design offered all individuals with the ability to review any quotes drawn directly from their transcript. This issue is explored further under the sub-heading “research- related ethical considerations”.  was describing a decision to reduce or eliminate funding, they said that they would still need to make the decision because it was usually a time-restricted situation that was imposed by someone with more authority. In short, they would go with the best information they had available to them at that time.  37 2.12 Data Analysis Plan QSR NVIVO software was used to store and partially analyze the data. Individual interview transcripts and collected documents as well as personal field notes and memos formed the basis for a qualitative descriptive analysis. Each health administrator transcript underwent a structural unit analysis. I then used Inspiration software to visually map conceptual associations across the dataset. Examples of these maps are provided in chapters four through six. Qualitative descriptive analysis (Sandelowski, 2000) was used to examine participant perspectives related to what makes a good health care system; their perspectives on the value of using tailored health care services to alter sub-population health outcomes; and processes that participants located within BC health authorities go through when they must make decisions to fund, reduce, or eliminate funding for tailored health care services. My analytical plan was intended to keep my descriptions close to the data but to also acknowledge the constructed nature of the research process. In other words, while aligning with Margarete Sandelowski’s definition of qualitative descriptive research, my intention was to spotlight the interpretive nature of the reflexive turn in my work (Fontana, 2003). In the results chapters, I highlight key contextual considerations and organize participant perspectives along a continuum of proposed strategies for change. I achieved these results through a two-step process. First I used Labov’s analysis of narrative structural  38 units (Riessman, 1993). It is considered non-linear in the sense that the narrative elements do not necessarily follow a set patterned order. The analysis was useful in unpacking the narrative structure to understand how the stories were organized, developed, and where and how they began and ended (Coffey & Atkinson, 1996). I started my analysis by reading and re-reading each interview transcript and locating the story line(s). Next, I prepared a brief description of the story and highlighted general impressions and unusual features (Bottorff et al., 2000). I then established the narrative’s orientation by asking who, what, when, and where questions, followed by a description of any complications that were established through narrative turning points; asked evaluative questions that established the potential relevance of the story’s meaning; documented the results by asking “what finally happened?”; and ended the analysis with a description of how the end of the story was resolved (Coffey & Atkinson, 1996). In addition, I graphed the storyline pictorially to represent the narrative’s temporal nature (Bottorff et al., 2000). Riessman (1993) observes that Labov’s structural unit analysis omits the relationship between the interview participant and the researcher thus ignoring the co-constructive process of meaning making. As demonstrated previously, I attended to this dynamic by examining parcels of transcript text that demonstrate this dynamic. In sum, during this first analytical step, I considered the following questions while reviewing interview transcripts:  39 • How was the structure of the narrative organized? • Why did the interview participants develop their stories in that specific way during the conversation with the researcher? • What taken-for-granted assumptions were employed by the speaker and the listener during the interview that made the stories sensible (Riessman, 1993)? • “How [am I] situated in the personal narratives [I] collect and analyze” (ibid., p. 61)? I used the structural unit analysis as an entry point into my interview transcript portion of my dataset. I became familiar with how participants made sense of their role and the difficult funding decisions that they are required to make on a regular basis. By approaching my research with an eye on the importance of narrative and storytelling, I gained an appreciation for study participants’ lived experiences. Once all individual transcripts were cleaned of all identifying markers, I then began to order my interpretations of the entire dataset through the use of thematic coding (Coffey & Atkinson 1996; Creswell, 1998; Morse & Field, 1995). Themes became stronger when a larger number of participants presented similar perspectives. However, in cases where one or a couple of participants passionately raised an issue significantly different from the other participants, I also took note. These responses became “surprises” or “outliers” and I used these responses to question what they could mean in relation to the other documented perspectives. I have combined these two analytical methods for very practical reasons. I found that while the participants were in fairly powerful positions and could freely refuse to participate, I provided them with the commitment to ensure that their participation was confidential. Given the limited number of senior health care administrators across the province combined with the fairly prominent  40 case examples that they chose to share, I came to understand that it would be fairly easy for well positioned readers to identify who participated. As such, I decided that it would be ethically questionable to produce a true narrative inquiry of participants lived experiences because it would require providing readers with large portions of text that identified contextual and temporal information. My results chapters present a combination of what I have termed “strategies for change” and thematic responses. I created these strategies for change by combining participant narrative similarities related to specific issues such as what participants believe makes a good health care system. I then placed these storylines on a continuum to show the range of perspectives. Similar to Max Weber’s ideal types (Doty & Glick, 1994), the stories are intended to show the range of participant perspectives. As an analytical device, the continuums of responses are useful in describing the elements associated with a series of coherent perspectives. Individual participant perspectives may cross more than one strategy for change, but are more closely associated with one story over others. To complement my use of strategies for change, I have also produced themes associated with my three results chapter topics. I examined what and how people spoke about the important aspects associated with a good health care system, the usefulness of tailored health care services, and how good decisions are made. This was particularly useful in examining important aspects of each of the three areas and texture and intensity with which these aspects cross my research dataset. Together, the way I entered the data and the way I  41 organized my results have allowed for an examination of analytical connections that provide novel understandings of how participating administrators in BC health authorities make sense of what makes a good health care system, interventions intended to better sub-population health outcomes, and how they go about making good decisions on the public’s behalf. The quotes that I have chosen to display throughout my results chapter were chosen with care. First, I have not assigned a code or pseudonym for each participant because I am concerned that readers could link up multiple quotes across results chapters and compromise participant confidentiality. Instead, I make clear distinctions to indicate to the reader that a quote comes from different participants. In addition, I selected quotes that most eloquently captured the nuances that crossed a variety of participant’s positions on the given issue. I paid particular attention to ensure that, where possible, these quotes were evenly distributed across participants, the four health authorities, and from both rural and urban experiences. 2.13 Documents Internal documents, reports, emails, and other similar documents were requested and collected in situations where these sources provided context or information related to the process where the health care administrator made their funding decision. Collecting and analyzing documents was supplemental to the one-on-one interviews. For the most part, these documents provided supporting evidence of interview participants’ perspectives, frequencies related to which decision-making inputs were identified, and insights into other issues that  42 were prominent during the decision-making process. For instance, one participant described a complex tripartite health service development initiative. The planning group’s meeting minutes and terms of reference allowed for a greater understanding of the conditions that each partner brought to the table prior to agreeing to participate in the service planning process. My dissertation document set also includes reports that were listed on the health region website. Administrators were asked to state whether non-publicly available documents could be directly quoted or merely used to inform the analysis. Notes were documented to this effect and agreements were honoured. Member-checking, committee involvement, and an audit trail of field notes, methodological, analytic, and reflexive memos have been used to signal the level of trustworthiness of the data and my interpretive analysis (Herzog, 2005; Riessman, 1993; Sandelowski, 1995; Smythe & Murray, 2000). Memos were used to record and reflect on my personal biases, thoughts, and feelings; methodological memos were used to track key decisions; and analytic memos were used to construct the findings and discussion section of the thesis (Strauss & Corbin, 1998). I also documented any general observations during each interview. This included observations about the location of the interview and participant reactions. Additional consideration of research evaluative criteria will be examined below under the subheading “research evaluation criteria.”  43 2.14 Research-Related Ethical Considerations Hardcopy-based data was stored in my office in a locked file cabinet and all computer files were password protected. Upon reviewing the transcript, I eliminated any names (and replaced them with pseudonyms) and identifying features that could breech the participant’s confidentiality. I kept a master list of participant names in a separate location on my computer locked with a password so that when contact was required, I knew how to reach the individual. For the purpose of this dissertation I have returned to any participant that I quoted in part or whole. I confirmed with the participants that they felt satisfied with the way in which I have tried to protect their privacy and that the way I used the quote aligned with its original spirit. Only one person requested minor changes to his/her quote and no participant challenged how I situated his/her quote in relation to my analysis. See Appendix B for a copy of the informed consent form. Traditional research ethics concerns related to protecting vulnerable populations are less of a concern with the population I interviewed; however, the way I proceeded with my analysis and displayed my results were more of a concern. I did not offer the study participants incentives to participate, nor were they in a situation where they could consider that their involvement would grant them potential access to additional benefits, such as is the case where an individual is invited to participate in a drug therapy study. The participants who qualified for inclusion in the study were in fairly powerful management positions, hence their decision to participate or refusal to participate was not likely to bring with it external  44 repercussions. In cases where an individual refused to participate, he/she was not contacted again. I also kept his/her decision in confidence. In summary, I endeavoured to document ethical considerations, informed consent negotiations, and a clear research-decision audit trail, in order to provide an explicit understanding of how and when participants were involved, how participants were able to negotiate their involvement, and how I produced my interpretations. As stated above in the data analysis section, I entered my data through the use of narrative inquiry analytical processes, but shifted my analysis to align more carefully with qualitative descriptive research to guarantee that the way that results were presented ensured participant confidentiality. 2.15 Research Evaluation Criteria Attention to assessing the quality of qualitative research has grown over the past decade (Alvesson & Sköldberg, 2000; Bottorff et al., 2000; Chenail, 1997; Clandinin & Connelly, 2000; Lincoln & Guba, 2000; Whittemore, Chase, & Mandle, 2001). Readers are advised that research results ought to be subjected to tests of validity. Whittemore, Chase, and Mandle (2001) argue that “[i]nvestigators have the responsibility to clearly state study validity threats, prioritized criteria, and specific techniques employed. Influencing this decision is the research question, study design, and philosophical stance of the investigator” (p.  529). Researchers are encouraged to: be reflexive; seek out alternative hypotheses; explore negative instances; examine bias; and look for ambiguities in the data. In addition, the research should be critical at all stages of the project since this may avert uncritical  45 verification and dogma. A way of increasing the integrity of the research results is for the researcher to do continual checks on the interpretation process (Whittemore, Chase & Mandle, 2001; Alvesson & Sköldberg, 2000). Similarly, Bottorff et al. (2000) suggest three criteria of assessing the trustworthiness of narrative research. These include considering whether the analysis is persuasive, whether the individual narrative accounts are coherent, and whether the presented interpretive analysis is transparent. Ronald Chenail (1997) argues that triangulation helps the researcher reflexively situate himself within the study and in relation to plausible interpretive accounts. My intentions related to using multiple data sources is not necessarily to create a coherent whole that forecloses alternate interpretations but rather to provide alternative perspectives on the same events in a way that may open multiple plausible interpretations for the reader. It is important to evaluate the entire research project because it provides the reader with a stronger understanding about how much truth-value should be associated with the submitted research analysis (Clandinin & Connelly, 2000). This arrangement would take into consideration the entire research process including the necessary interpretive turns that begin with the selection of the research question through to the production of research texts. As a result, my research design includes embedded processes for documenting significant thoughts and decisions. As previously stated, I tracked this through the use of an audit trail consisting of reflexive field notes, methodological, and analytic memos, and believe that I have addressed the criticality, integrity, and explicitness criteria. Participant quotes drawn  46 from the dataset have been used in chapters four-six to support my proposed interpretations, thus addressing credibility and vividness. In situations where I have used participant quotes, I have followed a member-checking procedure. This process was used to ensure that the most plausible co-constructed meanings from the many possible meanings that could be drawn from the interview holds truth-value to that participant. I have provided participants with all portions of the dissertation where their direct quotes have been used. In addition, I have provided them with any associated interpretations (Sandelowski, 1993). They were provided with an opportunity to comment on and or request correctives to my analysis in relation to their quote. Thoroughness and congruence have been established by ensuring that my research process is plumb and the selected dataset has the appropriate characteristics to allow me to answer my proposed research questions. Sensitivity and trustworthiness have been established through the examination of differing interpretive positions of participants, the research committee, and myself. In short, I have asked the following questions: is the analysis persuasive; is there global coherence (by connecting the story to the story-teller’s overall goal of the account — such as justifying how they made a good decision); and have I provided a transparent account of how my interpretations were made (Bottorff et al., 2000)? Lastly, I believe that the findings have gained credibility in situations where others find the analysis useful in understanding how decisions are seen to be made by administrators, what decision-inputs are seen as important, and how stake-holders are involved in decision-making processes (ibid.).  47 In summary, I began this chapter by documenting my epistemological and ontological beliefs in relation to the generation of research knowledge. I have outlined qualitative descriptive research as my chosen research method and described my dataset, tools, and analytical processes. I have attend to methodological limits and have examined how the qualitative research findings ought to be evaluated. In the next chapter I offer a literature review that assists in situating the research findings that will be presented in future chapters.  48 3 LITERATURE REVIEW In this chapter I provide a review of relevant literature that will offer a greater understanding of my dissertation findings described in later chapters. This review will provide the reader with a foundation in key governance-related considerations, a variety of substantive and procedural decision-making models, and the role of public participation in decision-making processes. As well, I introduce basic concepts and theories related to power. I end this chapter with a demographic description of the interview participants. 3.1 Situating the Research Topic in Relation to the Literature The need to examine health authority decision-making processes is driven by a larger substantive normative claim, often expressed in the areas of Science and Technology Studies (Wynne, 2001), deliberative democracy, and civil society (Dryzek, 2000; Gutmann & Thompson, 2004), namely, that the governance structures designed to protect the populace have not done a good job. Within Canada, health researchers have documented a significant increase in public concern that health care administrators are not making transparent and accountable decisions about the services that the public are provided in a way that proportionally weighs individual and collective rights (Abelson & Gauvin, 2006; Abelson & Gauvin, 2004a). In a partial response to this issue, health authorities are becoming more explicit about their decision-making processes. Recently health authority-specific research has been published on priority-setting frameworks in Alberta (Halma et al., 2004; Mitton &  49 Donaldson, 2002) and similar research is currently underway with Vancouver Island Health (C. Mitton, personal communication, August 23rd, 2006). In addition, Vancouver Coastal Health has developed a preliminary document that addresses ethical dilemmas associated with the allocation of scarce resources (Vancouver Coastal Health Richmond Health Services & Ethics Committee, 2002), the Provincial Health Services Authority has completed at least one planning cycle using an explicit funding priority setting tool (Mitton et al., 2006), and according to my interview participants, Fraser Health Authority now uses a six-point priority setting tool when considering cost reductions. Public participation and considerations about how relations of power flow through institutions and stakeholder groups weave through governance and decision-making literature. As such I first introduce a review of governance ethics followed by an examination of normative and descriptive decision-making models. I also describe a series of commonly identified judgmental heuristics. Next I briefly examine levels of public engagement and associated tools and assert that currently the requirement of health authorities in BC to engage the public in policy decision-making processes is a soft duty. I end this chapter with a discussion of how power can be conceptualized and examined. 3.2 Governance Ethics A normative ethical conception of good governance allows us to understand whether agencies, or in our case health authorities, have carried out their duties to an expected level. In this section, I provide a working definition of governance, followed by a discussion of  50 what makes governance good. Next, I describe principal/agent relations, which examines why individuals will constrain the advancement of their personal agendas and instead forward an organization’s mission (Buchanan, 1996). I will end this section by introducing a series of essential governance-related elements in order to answer the question “how do we know that good governance is taking place?” The main purpose of reviewing literature on governance is threefold: to act as a framework for understanding why various research participant perspectives are articulated during the research interviews, to appreciate why specific mechanisms are used in research participants decision-making processes, and to strengthen the case for stakeholder involvement in health authority policy development and monitoring.15 Governance attends to the structures for making organizational rather than individual decisions. These structures establish who has what sort of power to make certain decisions about specific issues (Kaler, 2002). The actors must be organized in order to systematically direct and control actions taken on behalf of the institution. Organizational governance extends beyond the Board of Directors and employees to also include all stakeholders who have a vested interest in the actions carried out in the health authority’s name, including patients, public(s), health care professionals, and government. Governance structures, then, provide mechanisms for ensuring that multiple stakeholders have the necessary venues to  15 The focus of this dissertation is to examine health authority decision-making perspectives rather than analyze whether good governance is taking place per se.  51 express their values and interests, work through their competing expectations, as well as attend to their legal rights and obligations (Commission on Social Determinants of Health, 2007). Structures are also established to manage organizational risks by setting limits and providing incentives at the individual and organizational level (Buchanan, 1996). Essential elements of governance “include: a clear mission; responsibility; accountability; transparency; stewardship; flexibility; succession; representation; and simplicity” (McDonald, 2000, p. 21). I offer the definition of governance provided by the United Nations Development Programme (UNDP) and quoted by the Commission on the Social Determinants of Health (2007): [the] (sic) system of values, policies and institutions by which society manages economic, political and social affairs through interactions within and among the state, civil society and private sector. It is the way a society organizes itself to make and implement decisions. It comprises the mechanisms and processes for citizens and groups to articulate their interests, mediate their differences and exercise their legal rights and obligations. It is the rules, institutions and practices that set limits and provide incentives for individuals, organizations and firms. Governance, includes its social, political and economic dimensions, operates at every level of human enterprise, be it the household, village, municipality, nation, region or globe. (p. 22) This definition clearly articulates the significant importance of the role and influence that external stakeholders have on institutional governance. It also makes clear the requirement for creating venues for bringing the public’s voice into the organization’s  52 actions, thus establishing the legitimate role for public participation in health authority decision-making processes. This examination of public participation will be further articulated below under the subheading ‘public engagement’. Extending Michael McDonald’s (2001) conception of research-related good governance to the health authority realm,16 we can say that it “involv[es] the translation of collective moral intentions into effective and accountable institutional actions” (ibid., pp. 3, 4). Collective moral intentions must meet appropriate moral standards, not just statements that are arbitrarily labelled “moral” (ibid.). Attention is directed towards the moral intentions of the public and on addressing how the organization can implement appropriate structures that take effective and appropriate action on the public’s behalf. Good governance requires trust established through effective accountability and effective oversight of at least two key good governance components: good institutional design and good institutional performance (ibid.). In short, there must be implicit and explicit evaluative criteria for ascertaining whether good governance has occurred. Similar to John Kaler (2002), McDonald’s conditions allow us to ask: Who is responsible for what aspects of the activities initiated on behalf of the health authority? What is intended and why is it intended? and last, What are the organizational barriers associated with implementing actions informed by collective  16 Both research involving humans and the delivery of health care services can be said to be a common good, must ensure the safety of the people participating in research or using a service, and ensure the proper stewardship of public funds.  53 morals (McDonald, 2001)? In addition, a normative line of questioning allows us to examine the ethical nature and limits to action carried out by the health authority and the people who work for or through it. Oversight functions of a governance structure occur at multiple levels or tiers, and are both formal and informal (McDonald, 2001; Wieland, 2001). It is not possible to eliminate all potential risks associated with conducting business through large organizations. Therefore, good governance allows the right balance of reflexive problem solving and attends to organizational risks but is not overly risk averse. Through the use of governance, people charged with oversight authority and the establishment of oversight mechanisms enable the monitoring and approval of the actions carried out by individuals on the ground. Oversight according to McDonald can occur indirectly through accreditation processes to ensure that appropriate actions are being carried out, and more directly through the use of incentives and disincentives (McDonald, 2001). Incentives can include formal recognition and financial awards, and disincentives can include performance reprimands and demotions. Health authority actors are assigned the right to carry out actions on behalf of the institution they work for through a variety of avenues, including legislation established through the Government of BC Health Authorities Act (1996) and through professional codes of ethics. At the health administrator level, there are no consistent management-level codes of ethics unless they belong to the Canadian College of Health Service Executives. While the administrator will bring their own profession’s code of ethics to their administration role, there is no guarantee that the individual belongs to an association with the regulatory  54 authority to provide any sort of incentives and disincentives associated with workplace conduct. While legislation and codes of ethics provide guiding principles and concrete limits to action, health authority policies and procedures provide day-to-day rules. Even within this assemblage of rules, incentives, and limits, administrators have significant flexibility in relation to their decision-making activities. Trusting administrators to align with ethical standards allows them to carry out their assigned duties. While trusting administrators to act according to ethical standards, health authorities must still guard against second order risks associated with accomplishing their goals (Buchanan, 1996). Individuals within the health authority structure may act in ways that are meant to evade detection and assignment of responsibility for the behaviours that created first order risks.17 Principal/agent theory posits that many agency risks are associated with a principal requiring agents to carry out the assigned work of the organization (ibid.). “Principals engage agents to perform tasks which they are unable to perform themselves, or which they find too costly or inconvenient to perform themselves” (ibid., p. 420). Buchanan suggests that establishing internal constraints such as auditing tools, assigning clear lines of  17 First order risks are risks that are not unique to organizations. These are risks that can just as easily be found between simple interactions such as between a doctor and patient. These are considered simple because an individual can carry out the duties assigned to them through professional or social standing. Second order risks, on the other hand, are more complex in that most goals must be obtained through several agents working towards a goal. With the increased number of involved agents comes decreased control of assigning blame for actions that have not been carried out properly (Buchanan, 1996).  55 accountability along with external incentives, alongside ethical commitments that articulate a commitment to reduce risks associated with institution actors (i.e. health authorities), work to mitigate second order risks in organizations (ibid.). The importance of health administrator responsibilities turns on duties owed and obligations. Administrators can be said to be the cause of health authority actions that affect patients, families, and the general public as a result of the changes they make to the health system. They can be said to have been responsible for bringing about a situation or failing to do so (Kaler, 2002). It is through responsibility for assigned duties that a person/organization can be rewarded or blamed for actions. Kaler (2002) holds that “[a]ccountability is fundamentally and primarily about controlling conduct and, in particular, preventing misconduct (p. 329).” It tells us about our conduct in relation to carrying out our responsibilities. Most people are more concerned with how badly people carry out their obligations and therefore more attention is assigned to “accountability” than “due recognition”. Transparency of actions along with accountability for decisions made allows the health authority to prove itself to those who are governed (Abelson & Gauvin, 2004a). In the ‘public engagement’ section I will examine health region public engagement and its relation to policy development. I will also expand on its relation to accountability and transparency. Good governance mechanisms allow us to assess whether those who are charged with making decisions on our behalf do so through legitimate, transparent, and accountable ways.  56 I have argued that patients, the public, and other stakeholders have a legitimate role associated with informing and monitoring how collective moral intentions are integrated into health system planning. I have also stated that health authority administrators ought to be expected to act upon the collective’s behalf rather than personal self interests. Through the use of principal/agent theory, I posited that health administrators are likely to adhere to collective health authority strategic directions through a series of incentives, disincentives, and adherence to ethical principles that express a commitment to limiting organizational risk. Lastly, I have opened a space for the discussion of patient, public, and other stakeholder involvement in health system planning. Before discussing public engagement in further detail, I will introduce literature on both formal and informal decision-making processes. How decisions are made within organizations is an important element of the governance process. 3.3 Decision Making Frameworks and Models Within any governance structure, there are several normative and descriptive-focused frameworks that assert how the decision-making process can be evaluated and how people make decisions. For instance, Daniels and Sabin (2002) have developed a normative model, which provides four evaluation conditions for legitimate limit setting processes. According to Daniels and Sabin’s formula, the following criteria must be addressed for a decision-making process to be considered legitimate: 1. The process must be public, including the grounds for making the decision;  57 2. “Fair minded”18 people must agree that the grounds for the decision are relevant and focus on a shared/common good; 3. The process must be subject to revision and appeal and the process for this step must meet the requirements of the first two conditions; and, 4. There must be evidence of regulations that uphold criteria one through three (Daniels & Sabin, 2002). Although Daniels and Sabin's justice-based ethical decision-making framework is useful, Gibson, Martin and Singer (2005) argue that it requires a fifth step called the “empowerment condition”. This corrective is intended to ensure that power imbalances between various stakeholders participating in the decision-making process are explicitly addressed. This could include procedural mechanisms such as voting, longer periods to deliberate, and clear articulation of each group and/or person's role and responsibilities in the decision-making process. Decision-analysis (Clemen & Reilly, 2001; Keeney, 1992; McDaniels, Gregory, & Fields, 1999; von Winterfeldt, 1992) and naturalistic decision models (Klein, 1998), are examples of descriptive decision-making frameworks. Below, I provide an overview of how decision-analysis is conducted. Next I describe informal or intuitive (naturalistic) decision- making frameworks. Judgemental heuristics relate and are influential to both types of decision-making frameworks and therefore I provide a review of heuristics in relation to this  18 “Fair minded” people are those who in principle seek to cooperate with others on terms they can jointly justify (Daniels & Sabin, 2002).  58 topic. I explore how the frameworks are used to explain decision-making styles and how they relate to good governance within health authorities.  Decision-analysis is a four19 to seven-step formal time-intensive framework for reducing complex problems into manageable parts. This framework can be used by individuals and/or groups. One of the innovative aspects of this framework that uses both qualitative and quantitative modelling processes, is that it explicitly acknowledges and includes stakeholder value judgements (Keeney, 1992). Figure 3.1 is extracted verbatim from Clemen & Reilly’s 2001 textbook on decision-analysis.  19 McDaniels, Gregory, and Fields (1999) provide a four-step process that captures Clemen and Reilly’s seven- step process as it relates to the importance of engaging stakeholder groups in the decision-making process. I have chosen to list all seven steps to ensure that a fair representation of decision-analysis is offered to the reader.  59 Figure 3.1: A Decision-Analysis Process Flow Chart  Source:  Clemen & Reilly, 2001 During the first two steps, the decision situation is identified and alternatives are explored. This is sometimes a difficult action as surface issues can hide the real problem from the decision-making group’s perspective. A decision situation is comprised of four key elements: values and objectives, decision to make, any uncertain events, and consequence (Clemen & Reilly, 2001). Values are defined as things that matter to the decision-maker and an objective is defined as a specific thing that the decision-maker wants to achieve during the decision- making exercise (Clemen & Reilly, 2001). Several objectives that are identified within a  60 decision situation can be connected. According to Clemen and Reilly (2001), an individual’s combined objectives make up his/her values. Values and the decision context must be considered together because it is the context that establishes which objectives will matter and will ultimately eliminate all other objectives. This is called the requisite decision model (ibid.). Ralph Keeney, the author of several seminal articles and books on the use of decision- analysis especially in relation to environmental planning, argues the opposite to Clemen and Reilly (Keeney, 1988; Keeney, 1992; Keeney & McDaniels, 1999). Keeney argues that during the decision situation framing, the decision-maker should first clarify what their values are and, based on these values, establish their objectives. Finally, the objectives will guide the generation of alternatives. All three authors, however, agree that creating as clear a decision situation as possible is of paramount importance when using the decision-analysis framework. Based on the established decision context and values, the decision-maker then identifies the elements of the decision to be made (Clemen & Reilly, 2001). Some parts of the decision may need to be made sooner than others given contextual considerations, and therefore decision-makers are encouraged to identify what must be immediately addressed. For example the decision-maker can ask: What options do I have to resolve this problem? Some of these alternatives will be specific (fund a particular health service), while others will be based on choosing a value (such as when deciding how much to fund). Regardless, the decision-maker is encouraged to always consider whether the status quo is an option (ibid.). The decision-maker may be aware of things that may come to bear on his/her decision  61 outcome, however he may not know how. These uncertainties can often be named and we can guess as to the possible ways the uncertain events may resolve (Clemen & Reilly, 2001). In these situations, the decision-maker is encouraged to include in his/her decision analysis the outcomes that will have an impact on at least one of his/her established objectives. In situations where there are multiple objectives, the consequences relate to what happens to an objective (Clemen & Reilly, 2001). For example, a consequence may be that a service is funded for people managing their diabetes in a community setting. The consequences may be good because fewer people are presenting at the emergency department (specific objective), but the consequences may be bad as well if the financial resources used to fund the service could have been used else where to save a greater number of lives. Step 3 involves creating a decision model. This is a three-step process that includes: selecting and structuring the values and objectives into fundamental objectives and means objectives with specific measurable objective accomplishments; structuring decision situation elements into a logical framework by using decision trees and influence diagrams; and ensuring that all elements of the decision model are clearly defined, that all alternatives are identified, all uncertain events are identified, and that all consequences can be measured (Clemen & Reilly, 2001). An objective that is well-defined has an object, a direction that is preferred, and a decision context (McDaniels, 2000b). When eliciting values and objectives, Clemen & Reilly suggest using the “Why is that important?” (WITI) test. It assists the individual in  62 “distinguishing between means and fundamental objectives and reveal connections among the objectives” (Clemen & Reilly, 2001, p. 48). Decision-makers are encouraged to apply this test until they feel confident that the reason they have given “just is important”. This usually indicates that the final response is a fundamental objective (ibid.). By sorting through this information, the decision-maker can establish fundamental ends, the means of achieving those ends, and the measures associated with knowing whether they have achieved their stated objectives (Keeney & McDaniels, 1999). In short, this becomes an objectives hierarchy. Each lower-level objective should explain or describe important parts of the higher-level objective. Figure 3.2 describes a fictitious objective hierarchy. Figure 3.2: Fundamental Objective Hierarchy Diagram*  *Fashioned after Keeney’s 1992 proposed structure Figure 3.3 provides a depiction of a basic decision tree. Decision trees and influence  63 diagrams are complementary. They can visually depict the entire decision model that includes the objectives, the identified decision, any uncertainties and the consequences (Clemen & Reilly, 2001). This can be very useful for accessing a bird’s eye view of the entire decision problem. Figure 3.3: Basic Decision tree* *Fashioned after Clemen & Reilly’s 2001 proposed structure This is a simple decision tree. It provides the building blocks for designing more complex trees. Three rules should be observed when developing a decision tree. First, only one choice/option that stems from the decision node can be made. Second, each chance node “must have branches that correspond to a set of mutually exclusive and collectively exhaustive outcomes. Mutually exclusive means that only one of the outcomes can happen …  64 Collectively exhaustive means that no other possibilities exist; one of the specified outcomes has to occur” (Clemen & Reilly, 2001, p. 70). Third, the tree must include all possible paths that a decision-maker may pursue. This includes all possible alternatives and outcomes related to chance events. In situations where there are multiple fundamental objectives, defining measurement is a necessary step. This means developing a suitable scale that defines each objective’s attributes. Scales can have natural attributes — where everyone can interpret similarly such as when we measure in the form of dollars, hours, or days, etc. — or constructed — where the attributes are context-specific (Keeney, 1992). In this case, a decision-maker could use a five-point scale that uses words instead of numbers. An example may include using the values “best”, “better”, “satisfactory”, “worse”, and “worst”. Each point would then require a clear definition. A means-ends objectives network can be used to assist the decision-maker in “identify[ing] missing objectives in both the objectives hierarchy and the objectives network” (Keeney, 1992, p. 69). Figure 3.4 provides an example of a means-ends objectives network. This figure is read from left to right. Notice that the objectives that are further left on the page are achieved by means objectives further to the right. Arrows show which means objective influences a higher order objective. Also notice that two of the middle objectives in the lightly shaded boxes do not have means objectives associated with them. This would alert the decision-maker to seek further information.  65 Figure 3.4: Means-Ends Objectives Network*  *Fashioned after Keeney’s 1992 proposed structure Step four becomes more technical and often requires the assistance of individuals trained in decision-analysis procedures. During this step the decision-maker begins to choose the best alternative based on the stated options. To do this, quantitative tools are used in situations where uncertainties exist and there are computer programs such as Decision Tools® to assist in this work (Clemen & Reilly, 2001). The decision-maker assigns values to  66 all alternatives and selects the alternative with the highest expected value. The expected value is the weighted average of a possible outcome (ibid.). Utility functions are also elicited. These represent the stakeholder’s perspective of how desirable a consequence is. This provides a useful starting point when discussing tradeoffs between options (McDaniels, 2000b). A sensitivity analysis is performed during step 5. This form of analysis can lead the decision-maker to ask whether s/he is trying to solve the right problem (Clemen & Reilly, 2001). By this point it should be clear, and if a problem arises, it signals that the decision situation requires attending to, and the process should begin again. Sensitivity analysis also assists the decision-maker in appraising their value judgments for their importance in relation to the overall decision. Space is also created to shift value weights to see how these alterations will alter the overall outcome choice (ibid.). During the final two steps, the decision-maker assesses whether further analysis is necessary (Clemen & Reilly, 2001). If his/her analysis does not produce adequate solutions, s/he may need to revisit his/her problem definition and this will once again require steps one through five to be followed. If however the analysis provides the decision-maker with solid solutions to his/her proposed decision situation, s/he would be in a position to implement his/her choice. Several benefits can be derived through the use of decision-analysis. Decision- analysis provides a structured process for addressing complex problems and can be used in  67 situations where there are many different stakeholder groups. It brings clarity to what stakeholders value and acknowledges that even expert knowledge holds subjective judgments. It builds these subjective judgments directly into the analysis through the value elicitation and weighting process. Lastly, it provides an explicit decision-trail. As evidenced by the lengthy description of the seven-step process, it is also clear that individuals or organizations interested in using this process must invest a significant amount of time and resources. In situations where the organization has limited experience with using this framework, a fair amount of training will be required for both the development of clear decision situations, values elicitation, and quantitative weighting of value preferences. A more likely scenario would see the organization or decision-maker hiring outside decision- analysis experts to guide the process. 3.4 Naturalistic Decision Models and Judgmental Heuristics Naturalistic decision models and judgmental heuristics relate to individual decision- making processes. The lessons from this area of research are useful to better understand why people do not follow rational decision-making processes like those defined above. They are also useful in understanding individual behaviours that are part of a group decision-making process. In this section, I provide a review of Gary Klein’s recognition-primed decision model. Lastly, I examine the role of common judgmental heuristics and emotions in relation to decision-making processes.  68 John Lavis (2006) argues that health care decision-makers work in fast–paced environments. Given the speed at which some decisions are made, formal decision-making models such as decision-analysis are not often used. Within these environments, decision- makers may rely on rapid cycles of decision-making such as the model offered by Gary Klein. His work relies on judgmental heuristics.20 Judgmental heuristics are short cuts in thinking that people use on an everyday basis. Small individual steps are combined and remembered as one large step. Over time, individuals are able to recognize a pattern of cues that alerts them to act in a certain way without having to think through each step (Klein, 1998). In health care settings, professionals use this strategy regularly. For example Klein describes research into how nurses can detect that a premature baby requires antibiotics for an infection long before tests can be taken and analyzed. This research found that nurses rely on a host of cues that were well known in the literature as well as cues that they learned from other nurses in the neonatal intensive care unit. The researchers found that while the nurses could not articulate how they knew that the baby required antibiotics, their assessments were often correct. The researchers concluded that the nurses were not relying on any single cue, but rather a pattern of subtle cues that signaled the need to administer antibiotics. Once a nurse learned the basic cues, she or he implicitly practiced identifying the cues until they become tacit. Small steps were combined into larger rules that required less rational  20 The study of intuition can be traced back to seminal work produced by Kahneman, Slovic, and Tversky (Böhm & Brun, 2007).  69 processing (ibid.). These larger steps are combined in different configurations based on the context. Klein’s (1998) model is called the recognition-primed decision model (RPDM). He writes that most people make decision in a very different way than those suggested by advocates of formal decision models like decision-analysis. Klein argues that people: • assess the situation to see if it is familiar rather than comparing between options; • evaluate options quickly by imagining how the action will be carried out rather than producing a formal analysis and comparison. Imagining the option, they can identify weaknesses and ultimately avoid them; • look for the first workable option rather than the best option;21 • generate and evaluate options one at a time rather than together for comparative purposes; and • implement and learn from the experience (Klein, 1998). Klein’s RPDM relies on establishing a decision situation. The situation narrows the decision- maker’s attention to a series of cues that assists him or her in distinguishing patterns. These patterns are said to activate “action scripts” (Klein, 2003). Action scripts are plausible actions that the decision-maker could take to resolve his/her decision situation (ibid.). The individual is said to mentally simulate the action script and apply it to his/her “mental models” (Klein,  21 Herbert Simon describes this later strategy satisficing (Payne & Bettman, 1992).  70 1998). Klein argues that mental models are based on an individual’s experience of how things associated with a decision situation and similar previous experiences work (Klein, 2003). For instance, if the decision-maker decided that it was important to remove barriers associated with participating in a program, s/he may draw on previous experiences that would provide him or her with critical information about the patterns of service delivery that have effectively increased participation. Mental models require the decision-maker to have past experiences to draw on in order for this model to work (ibid.). The more experiences that a person amasses, the higher the likelihood that s/he will have good mental models to evaluate the usefulness of the generated mental simulations. Malcolm Gladwell (2006) describes the ability to narrow in on the relevant information to make a good decision “thin slicing”. In these situations, more information is not helpful — in fact, it can be a hindrance. Gladwell claims that the unconscious is quicker at assessing a situation than the conscious part of the brain and that it is very difficult to articulate the unconscious (ibid.; see also Roediger, 1990). And this is where Gladwell and Klein align: individuals who have built up expertise in an area are able to quickly make good decisions, however they may not be able to articulate the decision-making components. This is paramount to intuitive or tacit knowledge. Max Bazerman (2002) provides a sobering caution to those explicitly relying on judgmental heuristics: there is a likelihood that heuristic bias may skew the individual’s process and outcome. Bazerman cites cognitive psychology research studies that examine how to address situations where cognitive bias may surface and suggests that it is possible to  71 learn how to correct or “debias” decisions where the individual relies on judgmental heuristics (ibid.). He argues that individuals can learn to identify when they have inappropriately applied a decision-making heuristic. Similar to Klein’s model, Bazerman (2002) argues that individuals can build expertise in accurately identifying heuristic pitfalls, model the likely outcome, and apply the heuristic if the outcome is considered appropriate. Of the small number of heuristics regularly used in intuitive decision-making (Kahneman & Tversky, 1996), three commonly identified heuristics include availability, representativeness, and anchoring and adjustment (Bazerman, 2002). The availability heuristic assesses the frequency, probability, or likely causes of an event by the degree to which instances or occurrences of that event are readily available in one’s memory. If the event evokes emotions and is vivid, easily imagined, and specific, it will be more available than its opposite (Bazerman, 2002). While this may help the decision-maker in some instances, it may skew his/her judgment in others. The representativeness heuristic occurs when people look for traits that an individual (or event or thing) may have that corresponds with previously formed stereotypes (ibid.). Lastly, the anchoring and adjustment heuristic occurs when someone takes historical precedents or random information and uses this information to anchor an initial value and adjust it to yield a final decision. It has the effect of narrowing down the possible options within which an individual will generate new perspectives on the topic (ibid.). Research focused on the use of judgment and the role of biases helps us understand their role in intuitive forecasting. It assists us in understanding the psychological processes  72 that allow us to make sense out of how perception and judgment work (Kahneman & Tversky, 1996). A central feature of this research includes the importance that framing can have in mental processing. Kahneman and Tversky argue that the decades of research that they have produced on this topic takes into consideration the role of context in the problem representation formulation and recognize that different problems will be addressed in different ways (ibid.). Another important aspect of this research provides us with an understanding of how emotions assist individuals during the decision-making process. Citing neuropsychological and cognitive psychological studies, Pfister and Böhm (2008) argue that emotions do not imply “irrationality” and they are not separate from rational cognitive processes. Rather than attempting to separate emotional responses from rational decision- making processes, Pfister and Böhm argue that attention should be directed towards whether the emotions drawn upon are actually appropriate for the issue at hand.  … the issue of rationality should be based on the validity of emotional evaluations rather than on formal coherence. If our emotional appraisals are appropriate, that is, if we fear what objectively is to be feared, and if we hopefully anticipate what will actually make us happy, then these emotions might be called rational. (Pfister and Böhm, 2008, p. 8) These conclusions lead Pfister and Böhm (2008) to propose a four-function framework of emotional mechanisms used in decision-making processes. First, a particular class of emotions can act as evaluative cues when the decision-maker needs personally relevant information when considering his/her options. Second, emotional responses and bodily reactions can accentuate the need to fit one’s decision-making process within the  73 context and timeframe allotted for the decision at hand (ibid.). Third, emotional responses can direct the individual’s attention to key decision contextual elements when appraising the decision situation (ibid.). Last, moral sentiments like guilt, anger, or love help the decision- maker hold to his/her conviction to carry out the chosen decision (ibid.). In this subsection I have examined literature that suggests that formal rational decision-making models are not always followed for a variety of contextual reasons – time and resources being a significant constraint. I have also shown that emotions and intuition ought not to be as feared in decision-making processes as they have historically been characterized. Decisions that first appear random are often (but not always) informed by past experiences, quick assessments of context and desired outcomes, and embodied emotional responses. Far from settled, the area of cognitive psychology continues to empirically examine how people in social settings draw on past memories in present time to affect future outcomes. They have, however, provided promising orientations for examining how decisions are made today. 3.5 Public Engagement As previously stated, one role of the public in relation to health authority governance is to articulate their collective moral interests, as well as monitor and evaluate whether the authorities have turned these interests into action. Health authorities have used public involvement mechanisms to involve stakeholders in policy discussions as a way to increase transparency, accountability, and legitimacy in the decisions that they make. However, these  74 activities have been met with varying levels of rigour and success (Abelson et al., 2003a; Frankish et al., 2002). I suggest that while the literature continues to grow in relation to this area (Chafe et al., 2007), overall, Canadian health authorities have used engagement activities sparsely and with different degrees of success. Although the government has conferred upon BC health authorities the responsibility to consult with the public (Government of British Columbia, 1996), this requirement was left open to interpretation and therefore there have been varying levels of commitment to this duty. Where some regions have consulted, others have created mechanisms to more fully engage the public through a continuum of involvement highlighting the level of involvement from one-way information sharing through to involving stakeholders in setting health system priorities. The MoH requirement could be considered a hard accountability imposed upon the health regions by government requirements, however, due to limited clarity and enforcement, it has acted more like a soft accountability, meaning the health authority actors choose the role that public involvement has in their decision-making processes. Clarifying whether the health regions are acting on hard or soft duties is useful in assessing whether the organizations actions are justifiable (Kaler, 2002). In relation to public involvement in health region decisions, then, it cannot be said that decisions are only justifiable based on whether or not they have engaged the public. Daniels and Sabin (2002) state that accountability for reasonableness is aimed at improving democratic deliberation and that “there is no general requirement that direct, participatory democracy has to play a role at every level within the institutions that make decisions (ibid., p. 166).” However, after reviewing transcripts of four  75 deliberative dialogues with 1600 randomly selected Canadians on a variety of topics, Abelson and Gauvin (2004a) found that Canadians want an active role in informing decisions that affect them. While it may not be a hard duty to involve the public at this point, this is likely to change in the future. In the meantime, community engagement can provide constructive functions by developing an understanding and conceptual apparatus to address challenging public issues (Kaler, 2002). Multiple organizing options exist for health authorities interested in institutionalizing public involvement processes throughout their structure. These include having community elected or government appointed health region Board of Governors, developing stakeholder engagement frameworks, assigning an administrator responsibility for leading health region community engagement, establishing a community engagement department, and establishing regional policies that outline engagement requirements in health authority decision-making processes. The proliferation of arguments for public participation has also brought with it a call for strong evidence of its effectiveness (Abelson et al., 2004; Abelson et al., 2003b; Abelson & Gauvin, 2006; Rowe & Frewer, 2005; Rowe, Marsh, & Frewer, 2004). In addition, it has become clear that the level of sophistication has increased in relation to engaging the public in governance decisions. Public engagement is quickly becoming a process of employing a series of structured and systematic activities to engage patients and citizens in governance decision-making processes at various levels of health organizations. Common themes that run through the engagement literature include power sharing (Fung, 2003); creating a  76 transparent process (Abelson et al., 2004; Abelson & Gauvin, 2004a); choosing between levels of engagement and selecting appropriate engagement mechanisms (Chafe et al., 2007; Consumer Focus Collaboration, 2000; Hariri, 2003); ensuring proper representation (Jewkes & Murcott, 1998; Mansbridge, 2000); and evaluating the process and outcome (Beierle, 1998; Fiorino, 1990; Rowe, Marsh, & Frewer, 2004). Within the normative frame of health authority public involvement, participants cannot influence administrators to invoke changes that place it in violation of the Canada Health Act, nor provincial legislation. The degrees to which health authorities can share their legislated powers are limited. At the end of the day, the health authorities must still be accountable for the activities carried out under their name. Power, however, can be shared in the sense of involving patients, care providers, and the public within policy planning processes. Deciding how to do this in a way that makes the process meaningful for all participants is paramount. Since Arnstein’s (1969) ladder of participation was introduced, researchers and practitioners have made distinctions between what each kind of engagement technique will produce, and the likelihood that stakeholders will be involved in making the decision under review. Levels of engagement range from stakeholders being passively provided with information, stakeholders providing information to set questions (consultation), and stakeholders participating in the design and implementation of the engagement process as well as participating in selecting the final decision (deliberations and ongoing committee involvement) (Rowe & Frewer, 2005). Figure 3.5 depicts a series of three categories of engagement and a listing of engagement mechanisms assigned to each.  77 Figure 3.5: Levels of Engagement and Associated Engagement Mechanisms Engagement Category Examples of Associated Tools Inform Consult Engage Newsletters & speeches One-time consultations & surveys Deliberative Dialogues & Citizen’s Juries While the degree to which the public was involved in decisions described by my study participants was not the main focus, the literature would suggest that engagement to some degree ought to be seen in BC health regions. During the interviews I asked each participant to describe any situations where stakeholders were involved in their process and to describe the mechanisms used to engage these stakeholders. In chapter seven I will draw on Rowe and Frewer’s engagement typology as well as associated engagement and deliberative democracy literature as I reflect on examples of the engagement found in my study transcripts. 3.6 Analyzing Power Bent Flyvbjerg (2001) provides an analysis of power that is guided by both Weberian-Dahlian (Max Weber and Robert Dahl) and Nietzschean-Foucaultian (Friedrich Nietzsche and Michael Foucault) interpretations of power. Flyvbjerg argues that Dahl and Steven Lukes, among many others in this tradition, see ‘power as entity’ (possession, sovereignty, possession, and control) whereas Michael Foucault “understands power in terms of its concrete application in strategies and tactics — power as force relations” (ibid., p. 116). Flyvbjerg suggests that drawing what he believes to be the best from both traditions will  78 produce a solid analytical process. To highlight the differences between ‘power as entity’ versus ‘power as force relations’, Flyvbjerg compares how Lukes’ (1974) 3D view of power and Foucault’s genealogical analysis would approach questions of power. In an analysis of power, Flyvbjerg argues that Lukes would privilege “who” and “where” questions that include: Who can adversely affect whose interests? Who can control whom? Who can obtain what? Who can secure the achievement of collective resources? Who is responsible for the outcomes of power? Who benefits from the outcomes of power? Where are the sources of change localized? Where are the points at which alternative arrangements or event could have made a significant difference? (Flyvbjerg, 2001, p. 116) According to Lukes (1974), this line of inquiry can assist us in understanding that how power erases alternate realities is one of the most powerful actions possible: [I]s it not the supreme and most insidious exercise of power to prevent people, to whatever degree, from having grievances by shaping their perceptions, cognitions and preferences in such a way that they accept their role in the existing order of things, either because they can see or imagine no alternative to it, or because they see it as natural and unchangeable, or because they value it as divinely ordained and beneficial? (p. 24) Foucault’s line of questioning, on the other hand, would start with small “how” questions (identified below) that prioritize the examination of details and concrete practices in everyday life.  79 Flyvbjerg (2001) describes an analysis of power that is guided by a concept of power that contains six features: 1. Power is both productive and positive as well as restrictive and negative (ibid.). 2. While power can be localized in centres such as institutions or individuals, it is seen as a dense constellation of omnipresent relations (ibid.). 3. Power is not stable. It is regularly appropriated and reappropriated and “exercised in a constant back-and-forth movement in relation to strength, tactics, and strategies” (ibid., p. 132). 4. Knowledge and power, truth and power, and rationality and power cannot be analytically uncoupled because they produce each other (ibid.; see also Alvesson and Sköldberg, 2000). 5. It is important to examine both the structure of power and the process. Power as entity theories are less likely to examine the processes of power. Drawing on power as force relations, Flyvbjerg suggests including the question “How is power exercised?” in combination with asking “Who has power and why do they have it?” (Flyvbjerg, 2001). 6. As a corrective to theories that focus on structures of power, Flyvbjerg suggests using Foucault’s attention to the small questions. These are the localized practices, or praxis. ‘How’ questions allow the researcher to examine the complex configuration of realities that can be erased by using ‘what’ and ‘where’ questions. The ‘how’ questions, it is argued, attend to the details and concrete practices that are necessary in understanding how power flows (ibid.). Aligning with Flyvbjerg, and paying close attention to situations where my interview participants describe structures and processes (relations) of power, I will consider the following questions:  80 What are the most immediate and the most local power relations operating, and how do they operate? How has the active exercise of power in the relations being investigated affected the possibilities for the further exercise of power, with the resulting reinforcement of certain power relations and the attenuation of others? How are power relations linked together, according to what logic and strategy? How have these relations made certain rationalities possible and others impossible, and how do the rationalities support or oppose the power relations? How can the games of power be played differently? (Flyvbjerg, 2001, p. 123) 3.7 Conclusion This chapter began with a review of overlapping areas of relevant literature that I proposed would be relevant in offering a greater understanding of my dissertation findings described in chapters four through seven. I introduced the normative concept of good governance and how assigned mechanisms can allow us to assess whether those who are charged with making decisions on our behalf do so through legitimate, transparent, and accountable ways. I argued that patients, the public, and other stakeholders have a legitimate role informing and monitoring how collective moral intentions are integrated into health system planning. Through the use of principal/agent theory, it is posited that health administrators are likely to adhere to collective health authority strategic directions through a series of incentives, disincentives, and adherence to ethical principals that express a commitment to limiting organizational risk.  In the third subsection, I introduced a series of descriptive and normative decision- making frameworks and associated considerations. Descriptive models included formal  81 decision-analysis, recognition-primed decision model, and a description of the most commonly used judgemental heuristics. I also introduced Daniel’s and Sabin’s normative model used for rationing limited resources entitled ‘accountability for reasonableness’. Decision-analysis provides a structured process for addressing complex problems and can be used in situations where there are many different stakeholder groups. It brings clarity to what stakeholders value and acknowledges that even expert knowledge holds subjective judgments, thus collapsing the public values versus expert rational objective facts dichotomy. It builds these subjective judgments directly into the analysis through the value elicitation and weighting process. Lastly, it provides an explicit decision-trail. Drawbacks to using this framework include the required expertise, time, and costs. Whereas decision-analysis relates to both individual and group decision-making processes, naturalistic decision models and judgmental heuristics focus on individual decision-making processes. An example of a naturalistic decision model is Gary Klein’s recognition-primed decision-model and I provide a review of this work. Next, I examined the role of common judgmental heuristics and the role of emotions in relation to decision-making processes. I concluded that decisions — especially those performed by experts — that first appear random are often informed by past experiences, quick assessments of context and desired outcomes, and embodied emotional responses. Public participation and the examination of power weave throughout the literature review section topics. References to engaging the public in decisions can be found in governance ethics, normative models of decision-making, cognitive psychology-based and  82 economics-based descriptive decision-making frameworks, as well as in the analysis of power relations. I have introduced Rowe and Frewer’s (2005) typology of engagement levels and mechanisms as a way to order the storied experiences that my research participants describe. Lastly, I introduced a framework for understanding what power is and how it is used. Drawing heavily on Flyvbjerg’s (2001) analysis of power that combines elements of two traditions — ‘power as entity’ and ‘power as force relations’ — I documented a series of questions that assist my examination of how power flows through my research narratives. I highlighted my particular interest in moments where my research participants describe how power is used by themselves, their organization, or members of the public who are advocating for a sub-population-specific service. I believe this strengthens the qualitative descriptive analysis that I provide in the following chapters. I now move to a description and discussion of my research findings in chapters four through seven. I start by providing interview participant demographic information. This is followed by a description of what participants identify as a good health care system. I then explore participant perceptions of the value of services and initiatives tailored to the unique needs of sub-populations with poor health outcomes. The focus then turns towards how participants describe their process for making funding decisions and the normative positions they ascribe to making ‘good decision’ with ‘good enough’ data.  83 4 PERSPECTIVES ON A GOOD HEALTH CARE SYSTEM 4.1 Introduction In the first of three results chapters, I begin by describing the dataset demographics. In the first section I show that participants have obtained relatively high positions of power within their organizations, as well as a significant amount of education and management experience. Normative claims about what constitutes a good health care system are shaped by a nation’s history, shift over time, and may vary across groups within a nation state. In the second section, I describe health authority administrator positions on what constitutes a good health care system. While I refer to provincial and federal level health system issues, they are provided in relation to the health authority level. In this manner, I also discuss participant perspectives about the value of health authorities as an important vehicle for health service planning and delivery. In the third section, I show that in general the participating administrators support the goals of the Canada Health Act and the Medicare program as well as a series of additional elements that participants assign to a good health care system. I also argue that research participants explicitly promote a series of bioethical and governance-related principles and  84 considerations that guide the tough decisions that they must consider on a regular basis.22 All participants stated that the health system (macro) and, by extension, service delivery systems (meso), need to change in order to align with their preferred visions. While a majority of interview participants support a similar vision of what constitutes a good health care system, they have differing perspectives about the degree to which the system must change and what it would take to achieve their desired future state. In the third section, I organize these diverging perspectives around two strategies for change that frame the research participants’ opinions about how to achieve a good health care system: (1) we need to tinker with the health care system, and (2) we need a complete health care system overhaul. I conclude this chapter with a series of insights that will be developed in the upcoming discussion chapter. Below, Figure 4.1 provides a visual representation of the current chapter.  22 I introduce these principles and considerations in this chapter and carry them into chapter six that examines participant descriptions of how they make decisions. I also examine this issue in the discussion chapter (seven).  85 Figure 4.1: Chapter structure  4.2 Demographics I interviewed twenty-four health authority administrators from four of the six existing BC health authorities. Participant characteristics are represented in Table 4.1. Significantly more women than men were interviewed (3:1). I found no noteworthy gender differences related to management role, work experience, or education (See Table 4.1 & 4.2). Of the twenty-four participants, a majority reported having a clinical-related Bachelors degree (N=17), most often in nursing (N=12). Twenty participants reported also having a Masters degree. In situations where the participant pursued a Masters degree it was most likely health care administration or business-related. The mean average of management level work experience across the dataset is 16.83 years with a mode of 20 years and a median of 16 years. Two plausible reasons for why more women than men would agree to be interviewed include (1) women are more likely to agree to participate in research studies (Daly, 1992), and (2) women are more likely to enter caring professions such as nursing. Since many of the participants described working as a clinician and then moving into management roles, it is not surprising that there are generally more women than men in health authority management  86 positions.  Eighteen of the twenty-four participants currently hold director roles or higher.23 Many of these individuals have worked in BC health regions since their inception in 1994 and several others were recruited from Alberta health authorities.  23 Health authority management levels are regulated by the Health Employers Association of British Columbia (HEABC). Senior management levels from lowest to highest levels are as follows: director, executive director, chief operating officer, vice-president, and chief executive officer.  87 Table 4.1: Select participant characteristics (N=24) Characteristic Category Count Gender Female Male 18 6 Highest level of Education Graduate degree Bachelors degree Other 20 3 1 Clinical training  Medical and allied health professions N = 17 *Most were Nurses (N=12) Total Female Male 6 5 1 4 3 1 3 2 1 9 7 2 Number of years management experience   5-9 10-14 15-19 20-29 30+ 2 1 1   88 Table 4.2: Employment Position Level Compared by Gender Position Female Male Total Vice President/ Chief Operating Officer 1 1 2 Executive Director 4 1 5 Director 8 3 11 Manager 4 1 5 Coordinator 1 0 1 Total   24 4.3 A Good Health Care System Figure 4.2: Elements  Participants were asked what makes a good health care system. I invited them to structure their responses in whatever way they saw fit. Where appropriate, I used prompting questions such as asking for general principles that would drive the system, what kind of  89 services would be offered, and what structures would be necessary to realize their version of a good health care system.24 Many of the participants chose to redirect the discussion towards what was currently working well or needed improving at the health authority level. As a result I found it difficult to separate participant perspectives on the Canadian health care system in general and the health authority level in specific. Many of the perspectives equally applied to both levels and therefore I have combined them throughout this chapter. Importantly, the perspectives described at the health authority level are complementary to the Canadian health system legislation and policy level. Interview participants provided a series of well-known bioethical principles that they argued ought to guide the health care system. Participants most often described these principles through the use of practical examples rather than naming them. The principles I identified participants describing included beneficence, non-maleficence, autonomy, and justice. Beneficence, a principle that states that one ought to do good, and non-maleficence a principle guiding people to do no harm, were raised in discussions about both publicly funded health care systems and direct patient/client care.25 The following quote comes from a  24 See Appendix A to review the interview protocol. 25 Health care professionals and administrators have additional responsibilities towards the public they serve when it comes to beneficence and non-maleficence. These duties extend beyond, for example, an individual’s duties to a stranger (Beauchamp & Childress, 2001).  90 participant who describes how professional and organizational standards help to keep patients safe. We all have professional organizations that help define what we need to do … Also there’s lots of things like the accreditation [where] they have their standards and expectations. And you know we have to keep the patients safe. We have to provide good care. There are those standards, and we need to kind of use them. Participants described how these standards are useful especially when there are organizational financial deficits. They help participants think through the impact of a proposed cost cutting decision. Autonomy was discussed in relation to providing care that is centred on patient/client- defined needs and ensures that they have a role in their own health care decision-making. Respecting the individual’s choices and inviting that person into the decision- making component of [care planning]. Enabling [the client] so that people are able to make those decisions if they want to live at risk … I guess it’s … sort of autonomy and decision-making. This participant, from a different health authority than the first quote, highlights a series of key elements associated with respecting a client’s autonomy as they interact with the health care system. It also demonstrates a shift in health care provision more aligned with a harm reduction approach. Rather than expecting clients to be passive recipients of care, they are expected to be active negotiators of the level and kind of care that they wish to receive, framed within the boundaries of what the health care system can provide.  91 As described in chapter one, justice relates to the fair distribution of both benefits and burdens within society. The justice principle includes equitability, service, appropriateness, and population needs-based considerations. It’s always interesting when we [redistribute resources] because people get nervous and they don’t know what to do about those changes and [it’s a] challenge for our organization, including our Board, and the Ministry to stand behind those decisions. But it’s hard not to. Because it’s an issue around why, how can we serve one part of a population, in a significantly different way than another? It’s just not right? So there’s an element of fairness, it’s a matter of distributing the current pool versus adding to the pool. And we’re explaining a change fairly soon, around how we fund residential care. Because we fund them all differently so the amount of care that’s been able to be provided at each facility is quite different. Based on their funding. But the client needs are largely the same. Why should their funding, and care provision be different? So we’re actually gonna have to reduce, some including our own facilities, to bring others up … But it’s that’s an element of fairness for that resident, who’s in the under funded facility. How do you justify that, if their needs are the same as the facility that’s over funded, or relatively over funded? As the above quote demonstrates, health authorities are responsible for funding many community care facilities, however many of the service contracts are inequitable due to historical funding agreements established prior to the health authorities’ conception. In their attempt to ensure that clients receive an equitable level of care regardless of where they live in the health authority geographic boundaries, this participant describes a formal process by which s/he draws on the justice principle to rationalize his/her administrative decision. Participants argued that a just health care system and services provided through it  92 ought to be effective and efficient.26 In order for the health care system and services to be equitable, efficient, and based on established needs rather than patient/client wants, participants provide other second order principles that they believe ought to be pursued. These include basing health system and service changes on empirical research and models that have been successful in other parts of the country or across the world. Participants also stated that services should be delivered close to where individuals live, in ways that are accessible to many different cultures and provided in multiple languages in culturally diverse areas of the province. When it comes to considering illness-based populations, the administrators argued that there ought to be an equitable distribution of resources to fund services across and within these populations. It was suggested that this might be achieved through a systems approach to planning across the continuum of care in order to establish priorities in a formal and fair process. While it could be argued that this approach already occurs to some degree in BC health authorities, participants are signalling that the current approach could benefit from additional changes. For instance, one participant questioned why cardiac services that serve so few people are funded significantly better than mental health services that provide care to a greater number of the population.27 This administrator goes on to passionately argue that  26 Accessibility weaves throughout several sections in this chapter, as well as through the future chapters, and for the purposes of this section will be set aside. 27 Questions of a utilitarian nature are raised throughout the results sections as well as the discussion chapter (seven).  93 when it comes time to fund or expand services in a health authority, there should be a formal process that takes a holistic approach to planning. In this way, the participant posited, the decisions would be more procedurally fair than current practice and potentially ensure that selected system and service changes are more likely based on population need. In a similar vein, another participant stated their concern that equity does not exist in the current health care system and once again calls for fair and just procedures towards addressing population needs. I’m quite concerned about the [population] inequalities. We still really don’t have the accountability filters or mechanisms to assure that we’re doing things in a way that allows everyone the same access. A different participant explains that helping marginalized people may be a socially important pursuit, but it also helps the health care system. I think that there are people who don’t have capacity [to use mainstream services] and I think we do have some responsibility to serve them … If there’s a need and if there’s an opportunity to do that, I think we do have some responsibility. But also it’s not all that pure [of a motivation] either. I mean it’s really nice if they’re staying at the urban clinic and getting what they need instead of coming to an emergency ward that’s already over crowded, that’s already very busy. And so if you have some mental health underlying issues, this is not the environment for you. There’s too many stimuli. And then they have to wait a long time and I just think there’s a better way of doing it. In chapter five, I expand on participant perspectives about the use of tailored services intended to address the unique needs of marginalized populations.  94 In the main, study participants supported the areas of service currently being provided under federal and provincial legislation and funding. However, participants also described six key areas that ought to shift in order to realize what they identify as a good health care system. Central to many of these proposed changes is their support of population health promotion. Table 4.3 provides a summary of each required change. These six key areas were:  95 Table 4.3: Proposed health system changes28 Proposed Change Description Increase primary health care resources  Strengthen interconnectedness of family physicians and their relationship with health authorities Increase the number of services and initiatives that address health promotion and prevention (e.g. well-baby clinics, healthy heart programs) Reduce acute care resources  Develop interventions within community- based service areas intended to reduce emergency department visits and hospital admissions Reduce the number of ineffective medical procedures for those with less than six months to live  If a medical procedure is unlikely to alter the patient’s palliative status, restrict service to ensuring the individual is comfortable and pain-managed Greater focus on elder care  Provide a net of services and supports for elders with attention to supporting their ability to stay in their homes as long as is safe Increased attention on chronic disease management Where appropriate, integrate chronic disease management activities within existing services Enhance service access through increased use of one-stop approach to services Consider which hospital based services can be moved to community health centre or “hub” settings Where appropriate, increase the amount of services provided at the community health centre (rather than the client’s home) Enhance service navigation The following quotes provide examples of what two administrators stated when asked  28 This table is not prioritized.  96 to describe their preferred health care system. So the world according to [me] would have, the patient, at the centre. There still is a need for acute care, but it would be balanced. And we’d be providing services that we need. And we’d get rid of the duplication of our location. There would be community-based care, there would be NGOs [nongovernmental organizations], and it would be all connected, and interfaced, by a robust IT [information technology] system, and there would be interdisciplinary care. Because one single provider can’t do it all. So there’d be a network of providers, engaged physicians, pro-active patient-focused, wellness-based. There’d be incentives for the management of chronic care and interdisciplinary care. There’d be attached patients. We’d be using clinical guidelines. We’d be looking at targets. It would be fully integrated, continuous and coordinated care. There would be data exchange and knowledge management, and self-management. Which is the biggest growth area in the next ten - fifteen years. To me, an effective system, an efficient system, a good system, would be one that’s grounded in inter-professional care and primary health care. A real strong primary health care system, with an interdisciplinary focus, that really enables, access to primary full-service, broad-service, primary health care services, meeting the basic health care needs of our constituents, of our population. I would actually see much less focus on [medical] technology and, acute care interventional work with most of those resources diverted towards health promotion and primary prevention — a lot of those resources. And I would also in all honesty, see that we would be rationing our acute care services to the people with the greatest need based on their quality of life. Based on their ability to live a meaningful life, because we don’t have resources that are boundless. But I think in the longer term we would have a much healthier population if we focused on primary prevention, determinants of health. Broad determinants of health, health promotion, [and] primary health care.  97 Combined, these succinct quotes provided by two participants from different health authorities capture the most salient elements of a good health care system raised by study participants. The administrators raise the importance of coordinated interdisciplinary care, evidence-based practice, prioritizing resources, working across sectors to include non-profit health and social service agencies, utilizing business support services such as a “robust information technology” system, and the need to shift resources from acute care towards “wellness-based” care provided in community settings. Both participants describe the importance of population health promotion. The first participant also provides wellness- based examples: a focus on chronic care and self-care management. 4.4 Benefits of health authority structure In general, participants were confident that health authorities are good structures through which to reform and deliver health care services to the public. The first benefit over previous structures includes their ability to plan across the health care continuum. Previously, all hospitals acted separately from community health services with each system having separate governance structures. While still difficult, participants argued that they are much closer to linking services across old system divides (i.e. primary health, long-term care, and acute services). Having responsibilities across the continuum have allowed administrators the ability to strategically plan and shift resources towards new priority areas. Being in a large organization also creates the economy of scale to realize savings in purchasing supplies and non-clinical services. Health authorities also have increased access  98 to additional non-clinical supports for clinical-based planning. These include business support, health system planning teams, statistical systems, and information technology systems and supports. Lastly, participants spoke of their ability to involve the public at the planning and policy development level of the organization. They saw value in engaging the public through these processes, however interview participant case examples and supplied documents show that in the main this is an area that is least developed for the four areas described in this subsection. Within the health authority structure, participants described their vision of what makes an administrator good at their job. Many of these descriptors overlap with descriptions of a good health care system, with some notable differences. I provide a brief description and expand this issue in chapter six that describes health care administrator decision-making processes. An examination of how participants describe what makes a good administrator shows that it requires drawing on both first and second order principles in a similar fashion described in the “good health care system” section. In addition, participants stated that a good administrator uses resources in a way that is wise, appropriate, and just. All participants argue that these practices are guided by epidemiological statistics, empirical research, and examining existing budgets. Participants stated that they do not believe that the best decision is always the most popular decision because they must often consider tradeoffs amongst  99 competing needs. As a result, they believe it is important to be accountable for their decisions. One way of achieving this is to leave a transparent decision-trail. While a good administrator acknowledges the importance of social justice, participants argue that s/he must balance those considerations with the needs of the entire population. As such, additional temporal and sociohistorical contextual factors are employed. A final attribute of a good administrator is that s/he coach others in his/her organizations. Participants identified the importance of grooming coworkers in order to support administrator level succession- planning efforts. In this section I have described normative conceptions of what constitutes a good health care system, and the administrators who work within them. I identified participant beliefs that health authority structures are good vehicles for altering the current health care system towards their preferred version. I also identified a series of six changes that participants would shift. In the next section I describe participating administrator perspectives on the Medicare program and additional participant identified priorities of a good health care system.  100 4.5 Medicare program and additional priorities Figure 4.3: Support for Medicare and additional priorities  Study participants identified principles and components of a good health system and health care service delivery system. They demonstrated general support for the five principles that govern provincial Medicare program funding transfers.29 In the main, participants simply stated their general support for the Medicare program. Participants consistently mentioned the importance of ensuring that services delivered at the health authority level were accessible to all citizens in need. This included services outside of those guaranteed by the Medicare program. Participant perspectives on accessibility are woven throughout this section.  29 The five program criteria include public administration, comprehensiveness, universality, portability, and accessibility.  101 In addition to citing strong support for the Medicare program, participants highlighted the importance of ensuring health system accountability and transparency, increased attention to primary health care, mental health, eldercare, and their capacity to self-advocate. I examine these perspectives below. 4.5.1 Accountability and transparency Accountability and transparency are seen by participants to be significantly important attributes of a good health care system. They relate to both system-level issues and individual administrator behaviours. The following two quotes accentuate the importance of administrator duties owed to taxpayers. The [public] needs to know that their dollars are being spent in a way that is the best for everybody. So that in fact dollars are not being wasted. Dollars are not being directed into services that either aren’t working or are the wrong services. So I think that there is a responsibility as well for us to have, to make sure that those dollars are being spent in the right way. You know you have to be accountable. And, accountability is a funny thing. In health care actually we’re not that accountable. And I think we need to put some more accountability into the system … I think it comes down to timelines and the need for change. So if your timelines are tight and your budget’s tight and there’s a need to shift the system, you have to be pretty clear about accountabilities. And what you are accountable for in that piece of the pie in the change. You know, for example, in many organizations in health care, if you go over budget it’s like “Oh well, there’s always next year, right?” So you’re not held to be accountable to manage within your budget. Cause you can always explain it off as a variance.  102 In the first quote the participant argues that health authority decisions ought to be accountable for the public dollars that they spend. However the quote also highlights that the public system is a common good and the system should be delivered in a way that is “best for everybody”. Importance is placed on getting this balance right and ensuring that services are effective. The second quote from a different participant speaks to both accountability and transparency. These proposed requirements are accentuated during times of tight budgets and short timelines and must have a clearly articulated reason for proposing system changes. The participant argues that administrators ought to be more accountable for their decisions and not merely “explain” away their failings. 4.5.2 Primary health care and mental health care Interview participants raised both primary health care and mental health care issues. Participants described the importance of continuing to pursue building stronger connections between family physicians30 and health authority services, supporting easier access to family physicians, and ensuring that there is a greater focus on health authority-lead activities that attend to chronic disease management as well as health promotion and education activities. The following quote describes the overlap between primary health care services and primary mental health service needs. Consistent and equitable access to services is once again  30 While family physicians bill the provincial government for services they provide, they are technically private business people. Many family physicians have little formal contact with health authorities except to refer a patient.  103 highlighted. I think that certainly within the mental health field, which is obviously what I’m most familiar with, there is something that is available in all communities across [identified city]. Do people have access? Yes, a cluster of people have access. Does everybody who needs mental health service have access? No. And I think that’s where our big gap is. We haven’t determined whether in fact it’s our system, or the primary health care system, and in the end if we were to be really client centred, they don’t care which system provides that service. What we know is that there is a gap in primary mental health care being provided. That it’s being provided inconsistently. It’s being provided and sometimes it’s not being provided. So there isn’t a formula for it and yet it’s the largest growing, health hazard that we have in terms of human wellness. And so what we’ve got now is a system that we recognize there’s a huge gap in. It’s getting bigger, and we don’t have a strategy to contain it. And we certainly don’t have a strategy that’s consistent across all of the health authority. What we’ve got is a number of projects that are out there that have kicked off. But it’s different in every health authority we go to. Or in every health service delivery area that we go to. Even across [identified city]. Although there are clearly differences across the city in terms of what people’s needs are and how people access service. The strategies that are going into place are very different and the delivery care systems are very different. So, with that, I think we don’t have consistency … that is not being well managed. So, I would say that we’re sixty percent there. I think that we do a really, really, good job with some things. And a less, adequate job with others. 4.5.3 Eldercare and capacity to self advocacy Participants described concerns about the impact an aging population will have on the health care system. They argued that as the baby-boomers grow old, there would be unprecedented pressures on the current health care system. They also raised the concern that  104 the health care system is difficult to navigate and the possibility that many seniors might fall through the service cracks. Below, an administrator returns to the Canada Health Act and links it to the actual delivery of care. In his/her discussion s/he raises the importance of eldercare and the ability of the seniors population to speak out for their needs in a time where cost pressures may limit their access to appropriate care. [The CHA principles], they’re overarching. They’re the things that we should have as our base values for delivering care in Canada. That assures that we’re gonna get that kind of care. And if we had those overarching principles, and we’re held accountable to them, we would deliver them. And again my example is the parent in your face is the best accountability structure there is. Because if you say [you’ll do] something, you have to do it. That’s what we need to help our senior population. I take my own mom and dad and their interactions with the health care system. And, they don’t live here, so I can’t advocate for them. I can’t be the one in there doing it, but they’re by themselves and they’ve just basically done what the doctors told them to do. And that’s the extent of their advocacy, self-advocacy. It’s inadequate. The participant weaves the importance of accountability throughout the quote. This administrator also implies a moral imperative that seniors ought to self-advocate for their own needs. The role of health authorities in assisting seniors to be better self-advocates, however is less clear. Even though the interview participants described supporting the Medicare program and population health promotion efforts, the degree to which they see the current system aligning with their preferred vision of a good health care system diverge. Some participants  105 believe that it’s only necessary to make minor adjustments to the system, whereas others believe that significant structural changes are required. In the next section I expand the discussion through the use of two strategies for change. 4.6 Health care system renewal strategies Figure 4.4: Two strategies  In this section I present two strategies for change crafted from participant perspectives on the extent of changes required to align the current health (macro) and health care service delivery (meso) systems with administrator-identified preferred conditions. The first strategy entitled “we need to tinker with the health care system” will be followed with “we need a complete health care system overhaul”. I arrived at these strategies by considering the entire set of transcripts and drawing upon the key elements related to how participants described a solution that would bring them to creating an ideal health care system based from their perspectives. Once I looked at these elements I examined which  106 ones were closely associated. For the most part, the strategies spoke to the intensity of how much change would be required. I then produced the different “strategies for change” based on this information. Sometimes people would express various elements of more than one strategy but the collection of elements from their individual interview aligned more closely with just one. These strategies are not mutually exclusive and participants’ perspectives overlap both frames. The two narratives, however, encompass the variety of presented participant perspectives. Both strategies for change share four similarities. First, budget lines have become tighter and health care workers are running at full capacity. Participants are clear that the current system cannot continue to grow at its current rate unless new resources are provided. Second, they accentuate the importance of integrating service planning across the care continuum and the importance of disease prevention and health promotion as a way to address health outcomes and health system cost pressures. [I]f you work well together [across the continuum of care] we could prevent some of the diseases that end up costing the system a lot in terms of acute days of care and so on. But it’s hard to invest in that, because the payoff for prevention is down the road and we normally work sort of fiscal year to fiscal year. So to invest heavily at the front end and prevent disease … if it’s not an immediate payoff to the politicians and you know the people that are funding, and so on, and there will never be, then it’s hard to invest in that. It’s easier to invest in increasing the number of beds, in a hospital, or increasing the number of beds in a prison. To contain at that end rather than to try to work at the front-end. I think that’s the biggest system problem that we have in any of these fields.  107 The above quote also speaks to the third similarity: the importance of shifting the system from one that mainly focuses on acute illnesses towards one that is mainly wellness-based. Put another way, the preferred participant-identified system is one that attends to enhancing people’s health rather than waiting until they are ill before providing service.31 This quote also highlights the difficulty of moving in their chosen direction in any significant way given the political nature of health care in Canada. Fourth and finally, participating administrators identified that many non-health system issues such as education, employment, and affordable and safe housing impact health differences and inequities. If we’re gonna talk about health we have to involve all kinds of different players at the table. You need to bring education and housing and the municipalities and all of the different government folks as well as the non-government folks to the table. Given that health authorities cannot directly or significantly influence these factors, this participant has identified the importance of participating in intersectoral partnerships focused on building strategies to address health differences and inequities. The good portion of the community depends on the partnerships that you have with others that influence [the social] determinants of health. So I needed to define health not by a health system but defined more [by] putting together the social context [of] health. The ability to create a healthy population would be the  31 Participants do not imply that acute services are not needed. However, they suggest that increasing funding for health promotion and education will ultimately reduce the acute care usage.  108 ideal approach. The health context needs to include aspects like housing, income support, education, municipal planning that supports the health of the citizens, etcetera. … [For example] how do you look from a whole municipality – what are all of the things it would take to help seniors stay healthy in that environment? All the way from, how do we design and construct the buildings? How do you design and make the roads? What kind of municipal policies do you have about taxation? All those aspects … How do you have a system that ingratiates all of those partners to create the right environmental context? The sustainability of the Canadian health system will not be dependent on how we design the “formal health system” but rather how we design and support the larger social and environmental context in which Canadian’s live. The latter quote accentuates the value of planning across and within systems. However, pursuing these expanded planning processes exponentially complicates the process. While participants know that this is an important direction to explore, how to do this effectively is far from settled. 4.6.1 We need to tinker with the health care system Those who advocated for “tinkering with the system” consider the Canadian system as a whole to be in good working order. They also hold that the system cannot continue to grow at its current rate without being provided with additional resources, and that the public must shift their expectations of what can be delivered unless they are willing to pay higher taxes. Well we have the principles of the Canada Health Act, which I think are pretty good in the main. We haven’t reconciled how we’re gonna pay for all that, which is an issue. Which is challenging us all, especially as the voters have voted to  109 really limit their contributions to the general tax base. But yet [they] still have a fairly significant demand and high expectations of the health system. So that hasn’t been reconciled. Nevertheless, participants suggest that it is possible that the health care system can be changed at a local health authority level in a way that, for the most part, attends to constituents’ needs. This can be achieved by using a system-wide approach to planning that relies on shifting health service delivery and budget priorities as well as additional dedicated long-term change management support dollars. Health regions are accustomed to receiving three year funding envelopes; however, this is not a long enough period to make a significant difference. In order to make an impact on health determinants, study participants argued that it requires seven to ten year funding commitments.  The recovered resources captured through shifting existing budgets and additional new change management funding are intended to cover the costs associated with piloting and stabilizing new program ideas that often rely on collaborations with non-health authority individuals and organizations. Central to this narrative is the belief that the bulk of health planning control should remain with the health authority even in situations where they are collaborating with community groups. A participant discusses the role of partnering with community groups when examining population needs. In many cases it requires working with other agencies to really understand what some of the drivers are as to why they get there. And then also focusing on, prevention and health promotion in those areas, so you can prevent those things from occurring in the first place. Where there’s not enough emphasis – there’s always some – there’s a lot of treatment call, which is fine. But you also need a prevention strategy.  110 Community groups provide health authorities with important information for their planning purposes. However, health authorities remain in the “authority” position – ultimately deciding what will be funded and how it will be delivered. 4.6.2 We need a complete health care system overhaul “I’m out to revolutionize the system” “It’s a whole culture change” The complete health care system overhaul strategy for change provides a much stronger claim that the entire Canadian health system must be completely overhauled in order to produce better aggregate health outcomes and ensure the system’s long-term viability for future generations. These stories are characterized by the need to pay more and better attention to marginalized populations; a reduction of unnecessary procedures; as well as a significant reduction in acute care budgets and a sizable expansion of primary health care services. Removing barriers so that marginalized populations can access health care services are framed as moral imperatives. One participant provides an ethical and economic argument to make a point: I think the system should be responsive and … accessible. And I think we should be removing barriers because we will not address inequalities, if we have barriers. So it is impossible for us to meet the mandate of the Canada Health Act unless we seek to do so. I also think that it is a moral and ethical responsibility to ensure that we are not preventing taxpayer’s access to services based on their  111 social, cultural, or economic status. So, I think there is an ethical-moral imperative. Deciding what ought to be included in a publicly funded health care system is relatively straight forward in some areas where most people agree, such as having access to emergency services, however, it is more difficult in the grey areas that require professional judgement and where people may have diverging perspectives. For instance, health care administrators and medical professionals must make case-specific decisions about which services are needed compared to those that are merely wanted by either the patient or their families. An administrator describes one such situation: When you see 80% of our acute care services are used by about 15% of our population, and out of that proportion, when I look at the futile interventions that take place in the last six months of life, and the expenses related to that, I actually think it’s futile. So when we have an eighty year old woman, with cancer, with multiple system failures, who is going to surgery for, something that’s not related to those specific diseases, and yet spending time in acute care and in intensive care because of her multi-system failure, for a surgery that’s not really aimed at improving the quality of her life, it’s futile to me. And, we have to wake up and, really look at what we’re doing. Significantly shifting planning priorities and associated budgets from acute services towards primary health care services and initiatives is another important element of the “overhaul” strategy for change. I would actually see much less focus on [medical] technology and acute care interventional work, with most of those resources diverted towards health promotion and, primary prevention. A lot of those resources. And I would also,  112 in all honesty, see that we would be rationing our acute care services to the people with the greatest need, based on their quality of life. Based on their ability to live a meaningful life. Because we don’t have resources that are boundless. But I think in the longer term we would have a much healthier population if we focused on primary prevention, determinants of health. Broad determinants of health, health promotion, [and] primary health care. In the above quote, the participant anchors his/her argument for making sizable changes to the health care system on the belief that there are limited available public dollars. This then leads him/her to open dialogue on what services ought to be provided and to prioritize who ought to be provided access. Reflecting on the nature of these philosophical discussions, several participants highlighted the importance of establishing meaningful dialogues with citizens about how best to pursue complete system overhaul changes.32 Finally, similar to the first strategy, collaboration with non-health authority individuals and groups is important but the emphasis in who should hold the planning power is shifted from the health authority towards community members and groups. This work is often described as ‘healthy communities’ collaborative planning. One of the most important pieces is that the idea for the program has come from the community. Not from a larger organization in the community, but you know, it just bubbled up … We began working with [a] rural [area], to develop an initiative where we would be able to fund three projects … I engaged community  32 Importantly, in discussing the need for significant system changes, participants still maintained that the Medicare program principles should be maintained.  113 developers in each of those areas. [The health authority] paid them for one day a week for six months to work with the communities in specific locations to identify a particular need, within obviously the areas that we could fund and to develop a proposal. To develop a community based advisory committee that would drive this, multi-sectoral as well. And this year we are funding the projects … serving vulnerable populations aligned with work that had already begun to move in those communities. While the health authority in this example continues to frame what can and cannot be funded based on their strategic directions, the community was able to shape what the initiative would ultimately become. This example also accentuates the importance of communities designing what would work in their context and how it would relate to other existing community-based arrangements. In this section I have provided two strategies for change that relate to how participants see achieving their vision of a good health care system. Both stories accentuate the importance of shifting towards a wellness-based system that uses an intersectoral and integrated approach to service planning. This approach makes disease prevention and health promotion central features in addressing long-term health care system cost pressures. The strategies diverge in relation to how drastic the required changes ought to be, and the importance they attribute to reshaping the relations of power between the health authorities and the communities they serve.  114 4.7 Conclusion Participating health care administrators envision a good health care system as one that supports the Medicare program, as well as governance ethics and bioethical guiding principles. These overarching principles are augmented by second order principles that help to shape health care delivery system service planning. Participants argued that a good health care system is logical, transparent, and accountable and they expect the same of health authority administrators. Some participants argue that health authorities may benefit from holding administrators more accountable for the decisions they make. Participants also stated support for ensuring that health care systems are transparent and accountable, and pay greater attention to primary health care, mental health, and eldercare. Next, I developed two strategies for change aligned with participant perspectives about how to achieve their preferred health care system vision. “We need to tinker with the health care system” and “We need a complete health care system overhaul” were used to highlight the similarities and differences associated with participant perspectives on how to reform the current system. Several issues have surfaced that will be pursued through the remaining chapters. The first issue related to administrator support for ensuring that services are accessible and for the importance of serving marginalized populations. Second, a tension was raised between the need for the public health care system to be socially just while ensuring that the needs of all citizens are addressed in an effective and efficient way. However, serving some segments of  115 the population may be costly given their additional service delivery needs. Third, participants identified the need to develop procedurally fair processes to guide funding decisions that require addressing tradeoffs between service needs. Fourth, participants identified the importance of engaging the public in program planning and policy development at the health authority level. Document analysis and interview participant transcripts indicate opportunities to strengthen this area. As such, chapter seven provides a broader discussion of health authority engagement processes. In the next research results-based chapter I examine in greater detail interview participant perspectives associated with the value of providing tailored health care services and initiatives to address access as well as health outcome differences and inequities.  116 5 PERSPECTIVES RELATED TO THE USE OF TAILORED HEALTH CARE SERVICES 5.1 Introduction Our society created and maintains unequal subgroups. And if the needs of the subgroup warrant a specific intervention that’s specific to them, let’s say that’s culturally appropriate to an Aboriginal, or an ethnic minority … that would be consistent with their teachings and practices, then I don’t see that as problematic. So my issues have to do with equality and respect at all levels. We have some clearly articulated health inequalities in this country and we definitely have some sub-populations with very, very poor health outcomes. And I have had experience working with programs that we often refer to as “tailored”, to the needs of the group. And I think that that is appropriate and health promoting to develop services that are accessible. And, that produce the kinds of outcomes that we’re looking for. It makes no sense to continue to deliver a service that doesn’t return value on our investment. Similar to the opening quotes provided by two participants from different health regions, study participants found tailored services to be a good strategy in the ongoing effort to close the health differences and inequities gap between populations. They cited a series of reasons for their support for this strategy: the health system is not designed to adequately service certain marginalized populations; regardless of the number of changes made to the health system, certain individuals will never be able to integrate into mainstream services; and lastly, this strategy satisfies justice, accessibility, and effectiveness criteria. Not all  117 participants, however, share this perspective. Those who are reluctant to support funding tailored health services cite the importance of reforming systems to ensure that marginalized populations can access mainstream services designed to serve all BC residents. These ideas will be expanded upon throughout the chapter. This chapter consists of seven sections. In section one, I begin with findings related to how health authorities become aware of the need for tailored health services and initiatives. In section two, I describe three strategies for change developed from participant transcripts that outline a continuum of perspectives in relation to the value of developing services and initiatives tailored to the unique needs of sub-populations and the degree to which they believe that mainstream services could ever address the needs of marginalized people. In section three I demonstrate that these strategies rely on two temporal considerations that frame the chances that health authorities will consider sub-population needs, namely, election cycles and when the request is raised during the year. Next, in section four, I move on to examine participant-identified factors that increase the chances that a service or initiative will be funded or eliminated and I show that once services or initiatives are funded, participants also describe a series of factors that protect them from being cut during budget crunches. These factors include the importance of the equitable distribution of services across the region, stakeholder engagement over an extended period of time, and the usefulness of formal evaluations. In section five, I describe a series of narrative complications related to the chapter’s topic and how participants resolve these issues in order to re-establish storyline equilibrium.  118 The three complications introduced include: (1) administrator vacillation between justice and utilitarian considerations; (2) stakeholder disagreements on issues and solutions; and (3) the way in which tailored services address equity and effectiveness but complicate system navigation. Prior to concluding, I provide an interpretation in section six of participant practices that support marginalized populations through small and dispersed micro-level actions. I develop four practices: (1) coaching community groups on how to build business cases; (2) taking external requests for services and stewarding them through the health authority system; (3) attending community-based meetings to build intersectoral partnerships; and (4) helping small non-profit and for-profit health and social service agencies build organizational capacity through coaching and seed33 contracts. Figure 5.1 provides a visual representation of the chapter. Each section will begin with the corresponding segment of the figure. Figure 5.1: Chapter outline   33 The term seed is used to describe the provision of a small amount of funding used to help program ideas germinate and grow.  119 5.2 Awareness of needs: How sub-population services get considered by health authorities Figure 5.2: Awareness of needs  Participants consistently described four common ways that requests for the development of tailored programs and initiatives come to the attention of health authority administrators. Not all entry points carry the same ability to see their requests approved and implemented by a health authority. Therefore I have ordered them from the most likely to be funded to the least likely to be funded.34  34 Funding for tailored services or initiatives can come from government and nongovernmental sources such as through donor agencies and individual gifts. The funds may be targeted for specific purposes or provided without direction. These funds could be drawn upon in any of the four cases described in this section.  120 5.2.1 Government orders Health regions receive provincial government direction in the form of a “priority list” or funding letter. The bureaucratic side of the government is obviously tied to the elected government of the day but can be considered separate from the second category. For instance, a single elected official may raise a one-time concern whereas the Ministry of Health may produce mid-range planning strategies (e.g. three years) announced by the government in power. In addition, an elected official may be from the opposition party, another level of government that does not directly fund the health authority. Two quotes from participants working for different health authorities highlight the political nature of the government forwarding particular sub-population needs. These quotes also highlight the participant cynicism related to how we currently select which services will be developed and initiated in BC health authorities. I think sometimes populations and clients get identified because they’re the political agenda of the day. Or [it’s] the flavour of the day or somebody’s pet project. I mean it’s a reality of where we work … I think those are usually identified for us in the Ministry of Health’s funding letter, right? Or those are mostly top-down decisions. And it’s a politically driven agenda. Politics plays a role. The Ministry of Health approach in BC in the last while is to identify some key health goals … And they have nine of them for this year including mental health and addictions. And they have seniors and wait lists and Aboriginal health on that list. And sometimes, they’re great goals. The Ministry of Health’s great health goals, or, maybe they’re the Premier’s great health goals, I don’t know. So that influences [health authorities], because often resources follow that.  121 Health regions receive Ministry of Health direction in at least two ways: (1) being provided with new targeted funding (usually one-time money or 3 year funding), or (2) being instructed to address the identified issue within the health authority’s existing funding envelope. When new money is provided for the proposed service or initiative, it may only be provided for a limited time, under the assumption that the changes can be absorbed into existing health authority budgets.35 Of the four ways that sub-population needs are brought to the attention of health authorities, government direction is by far the most influential when it comes to predicting whether the health authority will implement the proposed tailored health service or initiative. While it may be possible for health authority representatives to make cases for why they ought to not develop and/or deliver the requested service or initiative, it is not likely. This is especially the case since 2001 when the provincial government ended the practice of allowing the public to directly select who governed local health regions, opting instead to directly appoint all BC health authority Board of Directors.36  35 Generally, participants state that if they are provided with additional temporary resources, they will have time to restructure existing mainstream services and workflow and as such realize cost savings. They would then apply those saving to the newly established tailored service or initiative. In this case the new service or initiative would be considered sustainable. 36 Health Authority CEOs report to and are accountable to the Board of Directors. Board of Directors members are in turn directly accountable to the BC government for all health authority actions. It is fair to say that individuals who are appointed by the government are more likely to align with government policies over the issues raised directly by the public located within the health authority’s geographic boundaries.  122 5.2.2 Elected official influence Politicians, acting on the wishes of segments of their constituency base, pursue issues related to particular marginalized groups in cases where it is believed that tailored health services or initiatives may change that population’s current health-related circumstances. Participants argue that municipal, provincial, and federal elected officials will use their position of power to publicly state their memberships’ concerns as well as propose that the health authority ought to address the raised concerns in a timely manner. This can provide much needed help for one sub-population but as is often the case with choosing amongst funding options within a limited budget, it may be at the expense of another group in equal or more need. When I look at this province, we have a premier, who for whatever, reason, rightly, or wrongly, is focused on enhancing services in mental health and in Aboriginal people’s health, okay? So there’s a political person, who has substantial power and influence, that’s put this [issue] on all of our agendas. But then we have the Punjabi community over here. And we have the alcoholics over here. Then we have people in rural communities over here who are also often very underserved. How come their needs aren’t getting met? 5.2.3 Strong community/advocacy lobby Whoever has the most political power and influence will identify the people who will receive the more effective services. Sometimes health authorities become aware of the need for a tailored health service or initiative because businesses or neighbourhood residents raise enough concern through  123 health authority communication channels and public media. Groups that use this strategy may raise the need for tailored health services or initiatives because they are genuinely concerned for the marginalized group in question, or inconvenienced by having these individuals in their neighbourhood (i.e. active drug trade, prostitution, or homeless people). The [Business Association] was saying “We need [X] service because there’s so much public disorder on the street. We can’t run our businesses anymore”. In both instances, sub-population-related issues benefit from a sustained spotlight and the ensuing pressure that is applied to the health authority. As this next participant observes, the best organized and strong community lobby will not succeed without support from key advocates located within the health region at a management level. If the ground swell is from outside [e.g. community-based advocacy], and even if that ground swell has all the data to demonstrate success on many levels, that [proposal] still has to be sold internally to have the initiative supported in any way at all. Importantly, this quote once again demonstrates how health authority administrators must decide to fund certain services and initiatives over others due to limited budgets. 5.2.4 Health authority planning and management advocates Health authorities have begun to develop robust planning departments that support administrative decision-making processes, however, several participants informed me that they required easier access to statistical information. I would think overall in the system, I do think that we’re getting better at, at least  124 getting some of the information through [our] “health profiles”. So whether we’re getting it, through the census data, the stuff that the public health and the Medical Health Officers do, that show us the mortality and morbidity rates. The suicide rates, the incidences of certain diseases. And so, for example, we can start to take a look at “Gee, why do we have such a high incidence of diabetes in [city X]?” And then you start looking at, well okay, is it a specific ethnic group out there? “Gee we have a high rate of Aboriginal population there”. Or, “Gee there’s a lower income group there”. So you’re able to start kind of picking out and saying obviously there’s something going on where we have a higher incidence of something. Whether it’s a disease, or accidents or suicide. Whatever it happens to be. So we’re getting better at the data and we’re starting to use it. The planning departments provide assistance in gathering a variety of data37 such as epidemiological information extracted from health authority databases, and program modelling to examine how services would financially shift when variables such as provider- to-patient direct contact hours are altered. One of the largest challenges facing health authorities in relation to data collection is that each health authority operates several different databases and information does not flow across them. During 2001, when multiple health regions were amalgamated into larger authorities, the new health authorities inherited existing databases located in the hospitals and community health services areas. This dynamic creates barriers for the administrator when examining what would happen if service in one area of the health authority was reduced in  37 Decision-making data inputs will be described in more detail in chapter six.  125 order to provide additional service in another part of the system or to know whether an intervention resulted in a reduction of emergency room visits. In relation to unlinked databases, another participant states: That’s probably the biggest challenge. We can get good data on the health of the population and by program or service, but it’s very difficult to get good, credible data evidence that allows you or supports decision-making around moving funds from one area to another, or making a decision, or choosing investing in one area or another based on the outcome expected from the investment of that change. That’s probably our biggest challenge going forward. Well-placed advocates on the Board of Directors or employed by the health authority may forward the agenda of a sub-population from inside the organization. The advocate may have a personal interest in the issue, observed a consistent need based on empirical and anecdotal evidence, or chose to align with community advocates who have made a convincing argument. However, after the service or initiative is funded, its long-term viability may be threatened if the original internal advocate leaves the organization. In addition, selecting a service based on an internal advocate’s wishes can sometimes result in the service being placed in the wrong place and ultimately producing an inefficient service. Speaking to a particular case, another participant from a different health authority describes what happened when one administrator developed a service and then eventually left the organization. It leads the participant to list a series of questions that ought to be considered when developing tailored health services and initiatives. I think there’s risk in making things too specialized. I think … [services] are established by somebody who has an interest in that field. And what happens is  126 when that somebody wins the lottery, or leaves the system, or retires, or dies, or, whatever, all of a sudden you don’t have the champion that you need to sustain it. And so we end up with a lot of little programs tagging around that in fact don’t have the champion. Then somebody else has to pick up the gauntlet and run with it … When people created [the service], it was created because somebody had an interest. And it was [placed] in the wrong area … And I think that it’s always at risk of losing funding. Because its competition is systems that are way bigger than it is. And it is stigmatized and it is on the side … I think our challenge as administrators and people who develop programs, is to have some experience with [developing tailored services] and to always sit back and say to yourself “Does this really belong here”? “Is there somewhere else that it can be done better”? And the big question is “What is the sustainability of it”? Interestingly, one participant argues that the best-developed proposal will fail without internal management advocate(s) supporting their issue, while another cautions that this is not always good. Although internal advocates have the ability to influence the development of a tailored health service or initiative, s/he also risks developing a service or initiative that could be provided in the wrong way or place and without sustainable funds simply because they have control over that part of the health system. Both point to the importance of temporal, contextual, and historical considerations in relation to any decision that may be made. These considerations are developed further on in this chapter and in chapter six. In this section I have provided a series of participant interview quotes related to the value that participants ascribe to the strategy of developing health services and initiatives designed specifically for marginalized populations. I also described four ways that the need for tailored health services and initiatives come to a health authority’s attention: (1) elected  127 official influence; (2) government orders; (3) strong community/advocacy lobby; and (4) health authority planning and management advocates. These are not mutually exclusive categories and elements of several of them may be combined in a single situation to adequately explain how a particular tailored health service or initiative came to be funded. 5.3 Strategies Figure 5.3: Three strategies for change  I now further develop the range of participant perspectives related to the value of using tailored health services and initiatives to address access and health outcome differences and inequities. I produce three strategies for change and propose that they be read as a continuum of perspectives that capture all participant positions provided in the interview process: (1) the use of tailored health services is a good stopgap measure; (2) the use of tailored health services helps resolve issues of health equity; and (3) the use of tailored health  128 services is a poor use of limited resources. 5.3.1 The use of tailored health services is a good stopgap measure In “the use of tailored health services is a good stopgap measure”, participants hold that they support the development of tailored health care services and initiatives for people who are not using or responding well to the care provided by mainstream services. After all, as one participant states: I believe that our system as we know it, has been set up to serve a specific population for episodic care. So it’s really the working adults and what kind of episodic care they would need. That’s sort of how Medicare was sort of formed, right? So [for the] working person. The farmer that gets into a trauma on the field, has some sort of paid trauma and rehabilitative services, and then, back out you go, right? So, I don’t think that our systems are as accessible to Aboriginal [people and others]. While this strategy’s complication may be attributed to the inability of mainstream services to adequately attend to the needs of marginalized people, equilibrium is once again established through participant belief that the mainstream services can be altered to address the problem. This is due to participant convictions that the use of tailored health services and initiatives are time limited bridges that are employed until health authorities can successfully alter existing mainstream services to better meet the marginalized population’s needs. They emphasize the importance of changing the mainstream service delivery model as a whole  129 rather than acknowledge any unique needs that a given sub-population may have compared to the mainstream.38 Required service-related changes include training staff to become more culturally competent, as well as restructuring the existing delivery model in terms of how care is approached, the method that treatments are provided, and the location(s) that the service is provided in. I think the process [of developing tailored services] is worthwhile. I think with some of the programs we can now see if they’re filling a gap. That’s helpful, but as a system it’s not a system. You know it’s actually not a system. It’s a number of micro systems that aren’t connected … Some of the programs are filling in gaps. So obviously the benefit to the patients is worthwhile.  [D]espite the fact that we try to do things like the cultural competency and we have to learn about the sensitivities and how we can accommodate [people], will we need to have some streamlined or focused programs? I think we do. And I think that part of it will depend on what the barriers are to the system. Whether they happen to be a language barrier, whether they’re cultural barrier, whether it’s something else in the system that’s causing that barrier, right? So I do think, as much as I, think, oh my gosh, how many of these separate kind of entities are we gonna end up with? We probably need to be there, for a while. To get the attention that we need to remove those barriers. And hopefully, you know, ten years from now, fifteen years from now it doesn’t need to be, separate. It’s built in, it’s adjusted to reflect [the needs]. Remove those barriers. [emphasis mine] As outlined in previous chapters, my interview participants indicated that a good health care  38 In that respect, this strategy for change overlaps with the upcoming third strategy that argues that tailored health services and initiatives are a poor use of limited resources.  130 system would be one that was accessible, integrated, and easy to navigate. The first participant quote listed directly above proposes that while in the short term a tailored health service may be useful for “filling in gaps”, it comes at the price of maintaining a health system that is not really a system at all. In order to correct this concern and align it with their preferred vision of a good health care system, the tailored health service and initiative must ultimately be a short-term strategy while administrators redesign the continuum of health care services in a way that meets the needs of all constituents. When this work is completed, the tailored health services and initiatives will no longer be necessary and their associated budgets can be redeployed. As the second participant from a different health authority points out, it may be upwards of fifteen years before the necessary changes are complete. Other participants remained silent on how long they believe it would take for mainstream services to become truly accessible to sub-population needs. As one participant states, “I think that our system is very slow to change”. 5.3.2 The use of tailored health services helps resolve issues of health equity I actually think that’s the way to go, in many respects, right? Because otherwise you’ll never bring that sub-population up to comparable health outcome. You may never in any event for years, right? But I actually do endorse that. I mean to expect that sub-population that has poorer health outcomes for a variety of reasons – and we have a good sense what those reasons are. To expect them to access mainstream services, which would then improve their outcomes is, I think, naïve … You know [the maxim] “I can give you a fish, you’ll eat, for a day. I teach you to fish you’ll, eat forever”, right? I hate that saying. I just hate it. And I hate it because that approach is basically fine for young virile, intelligent, with no major barriers to living a decent life. Yeah you teach them to cook, you teach  131 them to shop, you know, the young kids. And so on. But to presume that for our eighty year olds and our people that have mental health issues. Or they have major health issues. I mean it’s hard enough for them to shuffle around in their little studios let alone to expect that they’re going to go out, even on these “Go” buses and pick up groceries and have all of the ingredients they need to come back and cook for themselves [is unrealistic]. A second strategy for change argues that developing tailored health services and initiatives are an important strategy towards addressing the needs of marginalized populations. They state that providing tailored health services and initiatives is not a stopgap measure as indicated in the first strategy, but rather, it’s the actual solution. Participants who follow this strategy acknowledge that the health care system makes efforts to maximize effectiveness and efficiency by providing a bulk of its services in a way that generally meets the needs of most people in society. However, this creates a situation where some marginalized people who have unique needs are not well served. Participants subscribing to this strategy for change believe that in order for the system to be fair and just, health differences that are inequitable as defined in the first chapter must be addressed in a systematic way of taking into account all population needs. As quoted in chapter four, one participant uses an ethical and economic argument to make his/her point: “It is a moral and ethical responsibility to ensure that we are not preventing taxpayers access to services based on their social, cultural, or economic status”. This strategy differs from the first one because participants argue that regardless of the radical changes made to the health care system (which they support pursuing),  132 mainstream services will never be able to meet the needs of the multitude of marginalized populations and therefore special services will always be required. Two reasons are cited: mainstream services are often delivered in a way that cannot easily integrate different worldviews thus leaving them ineffective, and many mainstream service users are uncomfortable mixing with some marginalized populations. To highlight the latter reason, one participant confides that many people using health service waiting rooms do not want their children sitting beside a drug-using homeless person. In addition, despite significant training on how to work with certain marginalized populations, some staff may not have the right attitude or belief system to handle all clients/patients fairly. The same participant goes on to say, “The care might be the same, but the delivery is very different”. 5.3.3 The use of tailored health services is a poor use of limited resources Tailored health services have been provided in every BC health authority. Even when the administrator did not support the use of these services, at one time or another they acknowledged participating in their design, implementation, and/or closure. A small number of participants suggested that there is limited proof that a tailored health service would make a significant health outcome difference for a given sub-population. Besides, they argue, everyone needs the same services regardless of who they are. Unique differences between populations such as cultural beliefs or practices, from their perspective, could be included in the process of providing medical care within mainstream services. One participant observed that tailored health services and initiatives are usually  133 provided in urban settings and rarely in rural and remote areas where there are often not enough mainstream services to go around. In urban, rural, and remote areas, participants state that there is not enough money to cover basic core services and therefore it is unwise to direct funds towards tailored services and initiatives. In short, specialized access is outweighed by generalized access. The same participant goes on to describe the development of sub-population services as “boutique” services — a term usually used to market small luxury apartments, hotel rooms, and designer shops. By placing the term “boutique” within the context of stretched health care budgets with more identified needs than resources to fund them, the connotation is clear: we cannot afford to fund activities that provide services for a small population at the risk of the majority. After all, existing services are available to all BC residents even if they choose not to use them. The participant concludes it is not a sustainable strategy. You have what has been sort of disparagingly referred to as “boutique programs” — small programs that do a very good job but are very focused, limited clientele. I don’t think you would easily question the quality of the service delivered because they are usually of high quality actually. Or perhaps even access, in which they’ve been developed because they have been developed in response to a particular need … Where it tends to fall down is around two issues. One is sustainability. Is a model of service which is very focused on a very small population really a sustainable model? … I think the second big knock on that approach would be affordability. Is it affordable across the whole population? … Is that really a good model when it’s just addressing this group? What is the whole need in the health authority?  134 Another concern associated with this third strategy for change is the belief that the health care system should be simplified. With the continual expansion of services and initiatives, the system becomes harder to navigate. As a result, the average service user as well as care provider struggles to find his/her way through the health service labyrinth. In addition, s/he must also understand any associated service admission criteria. Navigation is an important element of the study participants’ preferred health care system and this tension creates a different but equally concerning access barrier. This issue is reintroduced below under the complications section. A final posited concern is that developing tailored health services and initiatives may come as a direction from the government but at the expense of funding other needed services that have been identified through health authority planning processes and community advocacy. The following participant quote highlights the struggle associated with this arrangement. I’m of two minds about it. I think that there are a lot of services. There is a lot of focus on homelessness, right now, and there’s a lot of focus on crystal meth … issue based funding. The hot issue politically of the day, tends to get the funding — not that it’s not needed, but it gets funded at the expense of other populations, or other mainstream, if you want to call it, services that have been under-funded for years now. And Community Mental Health is one area where there’s been no new targeted funding to augment the capacity of mental health teams to provide follow-up. To expand their ability to deal with the dual diagnosed person. The substance abuse. The concurrent disorders. The homeless people, to provide the outreach. So, we get the funding around an issue, so crystal meth youth outreach. Great, I mean it’s really nice, but, really, crystal meth is such a little component  135 of the addictions issue. Why should we dedicate, so much, to such a little issue and ignore that, alcohol is such a bigger problem. And, meanwhile the addictions out-patient clinics are not getting much new funding for that generic counselling piece where most people [who] may have an alcohol problem might come for counselling … And it’s very frustrating, because, sure you’ll take the money and you’ll develop the service, but meanwhile the generic case management teams are scrambling and they’re overwhelmed. So that’s been you know, something frustrating to see. So I think it’s great to see that we [health authority] get the money, for the special population. But, um, we should not. As this participant explains, s/he struggles with maintaining existing services that are not adequately funded and the new resources that are made available to her team cannot be used for those purposes. His/her struggle relates to maintaining existing services while attempting to attend to additional areas that have traditionally received absolutely no support. In this section I have described three perspectives related to the value of providing tailored health care services and initiatives to address access and health outcome differences and inequities. These strategies for change frame the collective positions provided by the study participants. The three strategies make it clear that while many administrators may support the use of tailored health services to varying degrees, others are reluctant to use scarce resources in this way. Questions related to tradeoffs when choosing between who should be served and how the services should be provided are consistently raised by study participants across the three strategies. They raise the concern that there is no systematic way to choose amongst funding options that is just and equitable. Participants also consistently raised the political nature of the health care system in general and discussion of marginalized  136 population health needs in particular. These recurrent themes appear once again in chapter six and are examined in the results discussion chapter (seven). In addition to how tailored health services and initiatives get to the health authority’s attention, and what study participants think about the use of these services and initiatives as an effective strategy for addressing equity, access, and altering health outcomes, other issues come to bear on whether a service or initiative will be funded or eliminated. In the next section of this chapter, I describe a series of temporal, contextual, and socio-historical factors that frame administrator decisions. 5.4 Two significant considerations Figure 5.4: Two significant considerations  The interview participants in my study provide two significant considerations that  137 frame the likelihood that a health authority will examine marginalized population needs: election cycles and funding cycle timing. The former consideration relates to pressures outside of the health authority while internal organizational systemic processes influence the latter. 5.4.1 Election cycles Four-year election cycles create additional dynamics when it comes to the likelihood that a service will be funded or eliminated. While health authorities are at arms length from the political arm of the provincial government, they are nonetheless affected by political cycles. Participants described many instances where their decision-making processes were in some way influenced by politics in general and government funding announcements in particular. One participant highlights a consistent pattern that s/he has observed: “Funding is usually matching whatever time is left before the next election”. Another participant from a different health authority reflects on his/her long administrative career and argues: We know we’re gonna get more money [from the government] and we know we’re gonna do this. And we know we’re gonna do that. And yet, two years into the cycle, we know we’re gonna be cost cutting. Every single cycle of four years it’s been like that, in my experience in health care. It’s not good for the system. So it’s far too political. Generally, participants note that funding announcements flow during the build-up to an election and for the first half of the next four-year political cycle. This is followed by either a reduction of base funding or a requirement to maintain existing services without funding  138 increases. Once complete, the cycle is repeated. However, depending on the platform that the elected party wins with, the first two years of their term may be focused on fiscal restraint. 5.4.2 Funding cycle timing Funding cycles provide two additional indications of the likelihood that a health care service will be funded or eliminated. The first relates to the state of the entire health authority budget and the second attends to the point in time that the request is identified during the annual budgeting cycle. If a health authority is in a budget surplus situation, they are more likely to reinvest these unmarked dollars in identified priority areas. However, if the organization ends a fiscal year with a deficit, they are less likely to consider funding services that are not considered core services, services required by government legislation and funding letters.  Participants identified that health authorities begin their budget cycle planning at least four months prior to the new fiscal year.39 This is a large process whereby local Health Service Delivery Areas (HSDA) examine regional strategic directions and local priorities and propose budgets that meet multiple stakeholder-identified needs. Numerous opportunities exist within this four-month window for internal and external advocates to influence budgeting decisions. Beyond this four-month cycle, it becomes more difficult to influence the funding of a service or initiative unless new resources are identified. Often planning that  39 In BC health authorities start their new fiscal year on April 1.  139 involves stakeholders from inside and outside of the health authority will occur on a given issue from April to December in efforts to influence the following year’s budgeting cycle. One participant highlights how s/he strategically organizes his/her team’s future budget, coordinates funding priorities, and aligns all requests with organizational strategic directions. This participant refers to a planning document as s/he speaks. I create the big vision, right? Say for example in this one that I’m talking about [note: points to document]. So here’s my big vision. Then what I do is, the managers and I work through the high-level work plan. Then we attach the resources to it. And I know what my [projected funding] pot is and I can divide up my pot, how I feel needs to be divided up. I’ll attach resources to it. Then we break it out in chunks. Break it out into who’s doing what. And then it goes on to a Gantt chart. And then we hire the resources and then we know who’s accountable, for all of the pieces … So, it’s all tied in to this work, right? All the way from budget allocation to [implementation]. Now, this hasn’t taken me a day or two to do this, right. Because I knew how much I was getting. Then I had to do the work plan. And now my managers have to put in all the details and once I find out how much money [the senior executives] are gonna give me [I’ll firm up the budget] … If [a need is proposed but] doesn’t fit in [the department plan], like if somebody came to me and said, “Oh, gosh, I have a real interest in neuro- dynamics”, for example, it doesn’t fit [with department priorities], it doesn’t fit in anything. So it doesn’t even make it to the table. Right? So I would say “Sorry not appropriate for me. I’m not gonna consider that. You either have to go somewhere else or, to somebody else for extra funding, cause it’s not even gonna make the table”. It’s gotta fit within the direction and the framework that we’re using. This participant quote also highlights the embedded use of project management tools. S/he  140 has identified developing plans that are aligned with organizational strategic directions, assigning staff to develop plans, accountability structures, and using Gantt charts to visually represent progress on identified planning targets. These as well as other additional rhetorical devices will be examined further in chapters six and seven. 5.5 Predicting the likelihood of service selection and cuts Figure 5.5: Factors that protect services from cuts  Beyond the two aforementioned temporal considerations, there are also issues related to the coming together of several factors that strengthen the selection or cutting of a service or initiative.40 Participants highlighted that multiple stakeholder support for a similar issue raised through public and professional advocacy, combined with empirical research and convincing statistics, examples of proven interventions (a proposed plan), and senior  40 These factors apply equally to mainstream and tailored health services and initiatives.  141 management support for the proposed direction will increase the likelihood that this proposal will be funded. However, if a service or initiative has no proven model or success, lacks evidence of effective and efficient outcomes, and/or is considered too costly for the number of people served, the initiative or service is likely to be reduced or eliminated if it currently exists or ignored if it is proposed. Once a service is funded, participants identified a series of arrangements that may increase the chances that it will avoid being eliminated during periods where the health authority carries a deficit. First, tailored services that demonstrate regional equitability by providing the service throughout the organization’s geographic boundaries, and serve a significant number of people, are less likely to be eliminated, whereas unique services that are only provided in one or two areas of the health region and serve a small number of people stand a larger risk of having their funding removed. Second, there should be documented evidence that multiple stakeholder groups support the ongoing delivery of the service or initiative. This claim, participants argue, can be strengthened by showing that stakeholder groups were in some way part of the original decision-making through an engagement process that could include but is not limited to community consultations, participation on an advisory committee, reference group, and/or committee with decision-making authority. Third, after the service or initiative has been in operation, it is useful to show that there is community support stating that the proposed service model is the right model for their needs. Fourth, having a formal evaluation of the service or initiative that demonstrates an intended positive change on health outcomes, effectiveness, efficiency, and the use of expensive acute  142 services will significantly strengthen health authority support for the continuation of the service or initiative. 5.6 Complications Figure 5.6: Complications  Participant stories regarding the usefulness of providing tailored health care services in environments characterized by limited health care budgets highlight three narrative complications or plot turning points. The first relates to a standoff between philosophical positions, the second to interference, and the last to increasing system navigation complexity. Figure 5.6 highlights the three narrative turning points. Health care administrators are situated in a complex web of competing priorities and demands. Perhaps the most obvious tension raised through participant interviews was that administrators cannot escape the need to address utilitarian notions related to service  143 planning because there are limited resources (e.g. financial, physical plant, staff) and therefore are challenged to prioritize between competing needs. Principles of justice, equity, and accessibility may drive an administrator’s initial motivation to fund tailored services and initiatives; however, even in situations where the administrators supported the “health equity” strategy for change described near the beginning of this chapter, they indicated an organization-based need to consider maximizing outcomes for the resources spent or forecasting a significant savings in another part of the health region to offset tailored service costs. This arrangement was especially true in relation to considering the number of people served by a tailored service or initiative. Interestingly, utilitarian considerations were raised significantly less often by participants in relation to how effective the service or initiative was in reducing health outcome variances across populations. In instances where participants went beyond considering how many people would be served by the service or initiative and its ultimate draw on limited resources, to also consider its impact, participants described using crude proxy measures. The most often-cited measure was patient and staff perceptions collected through informal conversations and satisfaction surveys. The second complication relates to disagreements and interference with health authority planning processes. A majority of the study participants identified that five stakeholder groups weigh in on the promotion and prioritization of marginalized population needs: the Ministry of Health and government officials, health authority administrators, health care workers, members belonging to the marginalized population, and the general public. Conflict arises in situations where competing interests are not resolved. Each  144 stakeholder group holds a different kind of weight associated with influencing the health authority planning process, and each group exercises relations of power in different ways. For instance, participants identified that the government can direct the health authority to change planning directions, and health care professional associations as well as members of the marginalized population and the public can derail health authority planning by lobbying elected politicians and raising their concerns through media venues. Lastly, health authority senior managers can block support for an internally proposed service through limiting discussions about what can be considered for funding. One way this is achieved is by referring to organizational strategic directions and planning processes that have established service priorities. In chapter four, I provided a picture of what participating administrators viewed as a good health care system. In that chapter, they described valuing the importance of health system navigation. Producing sub-population services and initiatives appears to create a double-edged sword. While the services may increase access and provision of appropriate care, they also make the system more difficult to navigate. With increasing the number of available services comes additional rules about how and by whom these services can be accessed. Interview participants resolve the three narrative turning points presented in this section in subtle ways. Narrative equilibrium is re-established in the first instance through the use of judgmental heuristics that assist the administrator in deciding that the tailored service or initiative addresses the needs of “enough” people given the associated budget costs. The  145 second instance is resolved by drawing on public and other stakeholder engagement support to inform planning processes, and communication support to position health authority official positions and messaging, including any directional changes. The third complication regarding health system navigation leaves participants split between two positions. Those aligning more closely with “the use of tailored health services is a poor use of limited resources” strategy for change suggest that increasing the number and type of services unnecessarily complicates health system navigation. However, participants aligning with “the use of tailored health services is a good stopgap measure” and “the use of tailored health services helps resolve issues of health equity” prioritize issues of access, effectiveness, and equity over the possibility that the system may be more difficult to navigate. In the final section of this chapter I describe how administrators actively participate in micro-level activities that promote supporting the health needs of marginalized populations.  146 5.7 Making a difference in small ways: Figure 5.7: Four administrator-lead acts of resistance  According to the interview participants, budgets in BC health authorities are becoming less flexible. Administrators describe having less ability to freely shift resources as needs become identified throughout the year. Participants attribute budget inflexibility to two reasons: the government has increasingly required allocated dollars to be used in specific ways, and with tighter budgets comes less slippage to recover. In addition, attention to sub- population needs may be higher or lower on the organization’s priority list. Nevertheless, the participants in my study continued to find ongoing ways to support population health promotion initiatives designed to address the needs of marginalized people. While participants continue to work towards transforming the health care system at the health authority (meso) level through service redesigns, this is often a slow process. They describe supplementing this long-term planning work with everyday activities geared towards  147 promoting sub-population well being. I interpret these behaviours as micro-level acts of resistance intended to forward issues of equity in small ways. Figure 5.7 provides a visual summary of this section. Below I describe four ways that administrators support marginalized populations through small dispersed actions. Participants informed me that they coach groups on how to build a good case for funding: Tomorrow I’m speaking to [a community group]. And the question they’re asking about is why [X population’s] health is on the table in [this health region]. Because the other health authorities haven’t been able to put it on the table … Why do we have a unique structure for it here? And why do we have the capacity to advocate for that? The same participant then goes on to describe how to build a strong business case for his/her issue. Similarly, another participant from a different health authority informs a community group: We [the health authority] need to understand the population health needs. We need to understand the gaps in care. And we need to understand the drivers in health care that are going to be impacting the system in the next ten to fifteen years. And once we understand that, then we make some decisions. So if they could come to me in this example and say “You know we have the highest unemployment rate in youth, we have the highest drop out rates. We’ve got huge addiction rates. We’ve got this and that”, then I would say, “Okay, now you’re telling me something”. Once the administrator is supportive of the need, they bring the outside in by taking  148 external requests for funding and stewarding them within the system. This process can take the form of both informal conversations such as raising the issue during meetings, and more formal processes characterized by building comprehensive business cases and briefing notes as a way to build stronger senior executive support. Additionally, administrators participate in community meetings to help shape the community-identified issue and work alongside them to examine opportunities for co- funding projects. In these instances, the administrator works towards developing partnerships and presents themselves as just one funding body in an issue that crosses multiple sectors. Administrators also assist small agencies to build their organizational capacity. This includes providing the agency with seed money to initiate an intervention and/or building a stronger business case. These resources allow the agency to repackage their proposal in a format that the health authority is able to entertain. Resources may also be provided to the community agency in order to gather stronger evidence of a need or to evaluate success in addressing that need. As one participant states: We give out a lot of grants to different organizations like family counselling groups and all the rest of it. And really what you’re doing is, helping them to maintain their business. And helping them learn how to market or get investors. Or how to get donors and all that. So that’s a lot of these different companies, they’re small kind of “mom and pop” things. But [health authorities] wouldn’t survive without them. So, you just try to help them along. This participant is referencing the need to keep small for-profit and non-profit health and social service agencies afloat in order to provide services that the health authority has, for  149 various reasons, no interest in directly delivering.41 Examples include small long-term care facilities, addictions rehabilitation services, and homeless shelters. While larger formal planning process aligned with budget cycles may or may not demonstrate an appetite for supporting the needs for marginalized populations in any given year, the administrator maintains some autonomy to redirect small amounts of financial or capacity-building support towards a health and social service agency. 5.8 Conclusion I began this chapter by stating that a majority of study participants support the use of tailored health services and initiatives. However, I also showed that this support depends on how closely an administrator aligns on a continuum of three strategies for change related to the value of these tailored health services and initiatives. I have suggested that individuals who believe that tailored health services and initiatives are a good stopgap measure while they re-engineer the health care system, and those who believe these services and initiatives are the end solution, both support their use. However individuals further along the continuum doubt the value of these services and initiatives — opting rather to focus solely on reforming  41 Most notably, one participant identified that some services are too costly for the health authority to deliver because of union-related contract rules. In addition some models benefit from having peers provide the service or initiatives. While there are some examples where peers have been hired by the health authority to deliver a service, it has not been a quick or easy process for a variety of systemic reasons. As a result they have opted for providing contracted agencies with the resources to deliver peer-run services.  150 existing services so that they are accessible to all residents of BC. I then highlighted that election cycles and the time of year that a new service or initiative is requested helps to predict the possibility that a health authority may consider providing funding. Next, I showed that multiple stakeholder group support, combining empirical research and convincing statistics, proposing a proven service delivery model or intervention, securing health authority senior management support, and demonstrating that the proposal is both cost effective and efficient, combine to strengthen the chances that the proposal will be funded. Participants also provided a series of factors that predict the likelihood that an existing service or initiative will avoid being eliminated during years where health authorities experience budget deficits. These identified factors included the importance of the equitable distribution of services across the region, stakeholder engagement over an extended period of time, and the power of formal evaluations. In the third quarter of the chapter, I introduced three turning points in participant- aggregated narratives and explained how these complications were resolved in order to re- establish storyline equilibrium. First I suggested that participants often start from a position of supporting the elimination of inequitable health outcomes. However, participants state that they weigh these needs in relation to all other identified needs. Important deciding factors include the proposed solution’s affordability, and the number of people served, as well as whether the changes can be made within existing budgets. Second, legitimate disagreements can arise among stakeholders. Government officials are able to resolve discrepancies by directing health authorities to alter their proposed plans. Health care worker associations,  151 marginalized populations, and public groups are able to apply pressure on the health authority directly, through media action, and by lobbying the government directly. Health authorities respond to these pressures by employing communications support to craft strategic announcements and key messages as a way to stem the tide of dissent or show a shift in their planning direction. The third presented turning point related to participant broad-based support for making the health care system easier to navigate. Participant narratives however show that in efforts to make the system equitable and effective for marginalized populations, the associated proliferation of services and initiatives ultimately complicates system navigation. Participants who were less inclined to support the development of tailored health care services and initiatives use this argument as a reason for supporting their concerns, whereas administrators in support of tailored health services and initiatives perceive this to be a minimal issue as long as the tailored services are effectively linked into mainstream services. In the final section of the chapter I provided four examples where participants identified small ongoing actions that supported efforts to address inequitable health differences amongst the BC population. These activities occur alongside multiple year larger- scaled complex health system redesigns. The four practices included: coaching community groups on how to build business cases; taking external requests for services and stewarding them through the health authority system; attending community-based meetings to build intersectoral partnerships; and helping small non-profit and for-profit health and social service agencies build organizational capacity through coaching and seed contracts.  152 Three themes have emerged through this chapter that will be examined in more detail in future chapters. The first relates to how power is exercised at the health authority level in relation to the funding and elimination of tailored health care services. The second theme highlights the importance of health authorities taking a systems approach to health service planning in order to ensure that the tradeoffs associated with selecting one service or initiative over another are fair and just. This consideration leads us to the third theme: participants demonstrate that there are currently no satisfactory systems in place in BC health authorities to address competing health service needs. I now move the analysis to the third and final chapter that reports my research results. In the next chapter, I examine participant-identified decision-making practices.  153 6 DECISION-MAKING STRATEGIES 6.1 Introduction This chapter examines study participant descriptions of how they make health service-related funding decisions. I asked participating health authority administrators to provide information on the values that they ascribe to good decision-making processes, what key information they collect, and important clues used to evaluate their decisions. I also asked them to describe a situation where they made a decision to fund, reduce, or eliminate funding for a tailored health care service or initiative. During the interviews I asked participants to provide me with any documents that could provide additional information related to their decision-processes. In response I was informed of many dynamic and difficult decision situations. Participants provided me with a variety of documents that included but were not limited to meeting minutes, terms of reference documents, and internal organization briefing notes. Collectively, this document set provides the foundation for the last of three chapters outlining my research findings. This chapter consists of four sections. In the first section, I describe the values that participants ascribe to their decision-processes. I provide details on both formal and informal decision processes that the participants identified using on a regular basis. In addition, I offer participant-identified information on situations where they are more likely to use one decision style over the other. I end this section by recounting the decision inputs that study  154 participants draw on during decision processes. In the second section, I describe the tools that participants use during their decision- making processes. The tools create the conditions that help them organize their work processes, and strengthen health authority transparency, accountability, and legitimacy. In many instances, participants find these procedural tools useful however some participants find the formal processes restricting. In the third section I examine how participants come to believe that they made good decisions. I show that administrators rely on a mixture of temporally bound procedural and substantive indicators. These indicators relate to how the people came to their decisions, gut feelings, and whether anticipated decision outcomes were achieved over time. As in previous results chapters I conclude with a series of issues that will be further explored in the discussion chapter. Figure 6.1: Chapter outline   155 6.2 Framing the decision Figure 6.2: Framing the decision  Research participants argue that their decision processes are value-driven. In chapter four, I described how participating administrators identified a good health care system as one that provided accessible services for the entire population, and utilized resources responsibly in a way that was transparent and accountable to the public. Important guiding principles also included attending to effectiveness and appropriateness. As well, bioethical principles that included beneficence, non-maleficence, autonomy, and making justice-informed decisions based on needs rather than wants, were highlighted. In relation to justice, participants also raised the importance of making explicit the tradeoffs associated with attending to one issue over another and understanding the decision’s cascading impact on the entire health region. Collectively, these considerations frame participating administrator decision-making processes. In this chapter I expand on the justice principle by highlighting participant understandings of fairness and appropriateness. In addition, I introduce the participant-stated  156 importance of collaboration. These values underlie both formal and informal processes. Participants raised fairness in relation to the actual process of decision-making as well as where resources were directed regionally, and program-wise. Processes were seen to be fair when they followed a series of rules known to the decision-making group. The participants describe using a series of decision inputs and procedural tools that assist them in demonstrating decision-making fairness.42 The following quote describes a process used to evaluate funding options. It’s incredible when [you] typically do the scoring etcetera [of proposed funding options] and that list goes up, even though people have been passionate about their own priorities, the consensus is typically quite strong. But then we give some other folks that aren’t typically part of that process, because I’m talking about integrated health services now, service planning etcetera, to take a look at that again and form recommendations. So there’s a number of processes for people to have that considerate second thought, if you will. It’s not perfect, but so far it’s the best process that I’ve been able to find. This participant comments on a priority-setting process that relies on identifying priorities in a systematic planning process followed by a funding selection scoring exercise. The quote also highlights the value of increasing the number of people invited to participate in the exercise, but it does not identify a direct role in this process for consumers of service or the general public. The participant concludes with a comment on how the process has been an  42 These rhetorical devices will be described in greater detail in the next section.  157 improvement on past planning exercises, but that there is still opportunity for improvement. In addition, resource allocation decisions are seen to be fair when participants consider how limited resources are distributed across the region and program areas. A different participant from the same health authority as the previous participant describes how decision-making criteria helped the management team resolve the need to redistribute resources once it became clear through a planning process that existing distribution patterns were inequitable. So what they did was they looked at the needs of the population, and where the services matched — Who was getting what? What were the wait lists? And whatever, and then presented it [to the management team, health authority] wide, so then it became crystal clear you could see, you couldn’t stand away and say, hmm. So everybody in the [one part of the region] could just, kind of had to do that, look away, because you know, it’s black and white now. You can kind of see it. There’s a disproportion of waiting for services. So, can you live that way? When we’re all on one team working together? Well no, not really. So how do we start to move it along? The participant of this quote is also demonstrating the value of planning across a large geographic region. In this environment, it is more difficult for one city or town to continue to secure significant resources based on historical funding patterns at the expense of denying service to other areas of the region without demonstrating through a planning process that the need is still appropriate. Another dimension of appropriateness raised by study participants was how services  158 are designed and delivered. As discussed in previous chapters, participants described the importance of creating effective services that achieve anticipated outcomes for the populations they are intended to serve. We need to be constantly evaluating [whether] our services are meeting the needs of our population now. Let me take in a population that we wouldn’t typically discuss. A middle-aged white cardiac victim who has no experience with disclosing what’s going on in his personal life, with managing stress or anything related. Setting up a program where you just talk at somebody like that is not going to be useful. There will be no impact. We know health education does not deliver on its promise. There has to be an experiential and participatory opportunity for people to make the shift and integrate learning. So health education, which is talking at you, often a talking head process, has limited return. The participant providing this quote43 uses a mainstream population group to argue his/her case for tailoring services as well as for making existing services as effective as possible. S/he highlights the importance of using research and health system experience to inform the health service design. The participant implies that an effective and appropriate service is a good health service. A sizable number of substantive decisions within health regions are produced in team  43 This is a different participant from the past two quotes, but once again comes from the same health authority. While I have attempted to ensure that I provide quotes different health authorities within chapters, I found that the participants quoted in this section provided the most articulate understanding of the issue under discussion.  159 settings. Participants seldom spoke of making significant funding decisions on their own. Rather, it often requires discussions with several levels of management and occasionally Board of Directors approval. Not surprisingly, participants discussed the value of making important decisions in a collaborative fashion. I’ve been able to challenge people to step to a potentially more inclusive place, a more cooperative and collaborative place. I know we’re gonna do the kind of work we need to do to build community. As soon as people get into defensive models, I know I’m missing something. And I’m really referring to community process here. But for example within my own shop, I have asked each manager to contribute something of their budget to a common good. I have all the money elsewhere. But I don’t think it’s appropriate for me to harvest the money elsewhere. If they’re going to be collaborative, they have got to be truly collaborative. And I let them off the hook with the sense that, you know, bring what you can. But take an honest look at your budget. And I think this is really important. I can find the money, but I think it’s more important for us to find the money. And it’s that moving people out of silos into a connected and supportive environment, where priorities are shared. Burdens are shared, struggles are shared. And so, we then move forward in common, towards our goals. This participant quote highlights that collaboration is important for more than the immediate decision at hand. It also establishes the conditions for collectively moving forward towards a common vision. The participant asserts that burdens and struggles are shared as a way to communicate the importance of working as a team. In addition, the participant speaks about how s/he looks for cues to identify whether the collaborative work is on track. Observing “defensiveness” in the group participants acts as a trigger. The trigger helps the participant recognize that something was missed and must now be addressed in order for cooperation  160 and collaboration to continue. All participants identified situations where their decision-making process was either formal or informal. For the purposes of this dissertation, I characterize formal processes as ones that describe a series of structured somewhat linear procedures and tools that increase the likelihood that the final decision has left a decision-trail. Based on the decision-trail, an individual should be able to piece together how a decision was made, and how the identified stakeholders were involved in the process. For instance, through the development of a terms of reference document, meeting minutes, or progress reports, a formal process would likely identify who the health authority has agreed to involve and what kind of general structure would be used to arrive at a final list of recommendations and/or decisions. By contrast, I define informal process as those that may or may not document their decision-making experience, involve key stakeholders, or produce a decision trail. Participants identified several benefits for using both formal and informal decision- making processes. As health authorities have evolved in BC they have also become larger. With only six authorities covering the entire province, each is large enough to provide health authority administrators with business and decision support resources. Examples provided by participants include planning, communications, and risk management departments. Along with these departments, and aligned with business practices, come additional policies and procedures that administrators must follow. Participants identified a significant increase in the number of health authority employees that they are required to consult as well as documentation tools that they must use during resource allocation processes. Participants  161 welcomed the decision support and associated formal procedures. Benefits associated with formal decision processes that study participants identified include: increased transparency and accountability by creating a decision-trail through documentation (in various forms); organized process; clarity of who will be involved, at what point and how; how decisions will be made; and creating a higher likelihood that the produced decisions are procedurally fair. However, participants also identified that informal processes have the benefits of being much faster thus lending themselves to tight timelines, and increased flexibility by making it possible to effortlessly shift planning directions with less consultation. Participant-supplied descriptions of when they used formal or informal decision- processes lean on a series of considerations. I place these considerations on a continuum to show the shifting nature of variables that lead administrators to draw on either formal or informal decision styles. During both formal and informal decision-processes, the decision factors were weighted differently and appeared to shift depending on the decision situation.44  44 I came to understand the importance of these factors after analyzing my entire dataset, but cannot propose the conditions under which the various factors would be weighted in any given situation.  162 Table 6.1: Factors that influence decision-style  Formal Decision Process Factor Informal Decision Process Longer timeframe  Time  Short time frame (e.g. government imposed deadlines and daily expanding deficit) Significant redistribution of resources across programs or geographic areas Resource redistribution  Minimal or no resource reassignment outside of program or geographic area Divided Stakeholder positions Significant overlap No experience Proposed model Established local knowledge Politically contentious (i.e. with public, consumers and their families, provincial government or medical professionals) Degree of risk Minimal concern expressed by stakeholder groups Based on participant provided narratives and case examples, formal processes are most likely to be employed in situations where the participant deems it necessary to involve multiple stakeholders, most often requiring longer timelines (i.e. months to years), and where these stakeholders are divided on the issue at hand. Additionally, situations where it is likely that resources are being redistributed across programs or the region, where none of the participating decision-makers have experience with a proposed direction, and especially where the issues are politically contentious, also increase the probability that a formal decision process will be undertaken.  163 Not all processes require or are suitable for formal decision-making procedures. I identified at least four situations where participants stated that they were using informal processes. First, informal processes are more likely to be used when the organization has a sizable deficit that is increasing exponentially on a daily basis and requires swift action to correct the trend. Each day the health authority does not alter the spending course, there will be a higher likelihood that other priority areas will not be funded. As one participant states: “We had a large deficit that we had to deal with. We had to get it in line and under control and we had a short period of time to make decisions.” In these situations, administrators are also less inclined to engage many stakeholders beyond casual conversation. A second consideration relates to organizational capacity. While the health authorities are large, employees have competing draws on their time. Participants spoke about working in a fast paced environment and the need to prioritize how they spend their time. Any organization will have limits on how many formal planning and decision-making processes can occur at any given point. The first resource allocation tradeoff occurs at the point that a formal decision-making process is chosen. If study participants, and the host of other actors, are required to participate in one formal process, they may become unable to address another priority area. This selection process may be implicit or explicit. Third, informal processes align well with situations where a majority of stakeholders believe that one or more plausible options exist and one of these proposed solutions is the “right way to go”. In these cases, participants describe a collectively felt sense that the best option is obvious. Fourth, aligning with minimal stakeholder concerns are situations where  164 the participant intuitively senses that there is enough reliable evidence to establish an appropriate decision. This includes the identification of proposed service delivery models or solutions that the participant and/or staff have past experience in using. In other words, the proposed solution makes common sense. Here a participant speaks about a community development process where s/he comes to understand that there is an isolated group of adults with a variety of illnesses in one building who would benefit from being provided social services that would lead to accessing medical services: It’s just intuitive you know? It’s worth the risk of being able to go in and provide [the service in their building]. This is a group that otherwise is going to continue to be isolated. They will become more of a burden in terms of the health system. In these kinds of situations, the participant would also monitor the progress associated with the chosen direction and make course corrections on an as-needed basis. Course corrections depend on changing presenting needs and may include, for example, reducing nursing provisions and increasing pharmacy support. When describing common sense solutions, participants highlighted the importance of drawing on past experiences that are similar to the current situation. As such they appear to align with Gary Klein’s Recognition- Primed Decision Model. I discuss this alignment in further detail in the next chapter. In their pursuit of designing a more efficient, effective, safer, and equitable health care system, all participants described key decision inputs that they collect and draw upon through formal and informal decision-making processes. Collectively, the identified decision inputs include: geographical and identity/illness-based health outcomes and access to  165 services variance, and ways to either recover costs in order to reduce deficits, divest in services that are not health authority priorities (such as pre-employment programs), and/or reinvest in other priority areas. In order to achieve these goals, participants identified seeking out quantitative statistics through epidemiological data and demographic information, service utilization data, and qualitative data captured through key stakeholders with established experience in the area of proposed change, experience with various promising practices, and data supplied through stakeholder feedback mechanisms. Participants also identified producing and drawing on existing needs assessments from within their health authority as well as using existing assessments from outside agencies. These needs assessments often provide a combination of both quantitative and qualitative data. Participants also consistently spoke of consulting financial data to ensure that any changes they consider are plausible.45 Participants also spoke of using benchmarks or established standards of care that they can compare their service configurations against and examining other successful service delivery models during their health system service-related decision-making processes. In  45 During large-scale health service redesign processes, participants also spoke of producing business cases that examine the financial resources required to implement their proposed changes. Costs often surpass existing budget allocations. As a result, administrators must make a case for redistributing resources from other areas of the health authority system and demonstrate how the proposed changes will reduce costs elsewhere in the system. A consistent example supplied by participants relates to making changes in community-based services with the intention of reducing emergency department congestion, hospital admissions, and inappropriate length of stays in hospital beds as a result of the patient’s inability to access the proper level of care such as a long-term care facility.  166 addition, they consistently stated using peer reviewed research, grey literature, government and technical documents, and conference presentations. Often health care administrators have networks of people they call upon to seek information or suggestions. One participant states: You get to know your colleagues in [area in another HSDA], or, I can call on people from [another province]. So people have their groups that they can rely on … It’s a network. And the health care [network] is not that big in Canada. This participant has built far-reaching relationships with other health care administrators and researchers throughout the country through past employment, and attending conferences. Others also spoke of building these relationships by participating in health service research projects alongside academics. During the decision-making process, capacity-related issues are also examined. For example, participants spoke of examining the staffing skills that they required to address service redesign processes. This is distinct from examining the roles required. By exploring what skills are required, they uncouple the skills from the service provider role, which allows the administrator to develop new roles and human resource job classifications. In addition, participants described taking note of all resources available in their current context. By doing so, participants then frame practical options for solving their problem situation and thus narrowing the frame related to what options will be seriously considered. Examples include existing facilities, employees, expertise, and technology. Stakeholder feedback comes in many forms. While participants identified including non-health authority participants in their decision-making processes, others limited their  167 substantive decision-making processes to involving internal administrators. In both instances, however, participants acknowledged identifying key stakeholder groups that they believe ought to be engaged to inform the decision at hand — however the degree of involvement associated with this engagement spanned a spectrum of activities from informing stakeholders of their decisions, consulting about the decisions at hand, and involving stakeholders to some degree in the decision-making process. Participants stated that they captured stakeholder feedback through community consultation reports produced by the health authority or other community groups, through client satisfaction surveys, and/or through staff feedback. Feedback mechanisms are also embedded in organization-wide accreditation processes. Services seek feedback from their clients on a regular basis. This information is used at the service level to examine improvement opportunities, but it is also used in larger health region planning processes when appropriate. In larger formal processes with longer planning timelines, detailed reports are occasionally developed on identified issues and members of the public were invited to provide feedback. In situations where restructuring of services will be hard to reverse, proposed models are treated to extensive and detailed planning. During these processes, reconfigured services are mapped out through costing scenarios that consider different ways to provide direct care including the kind of required care provider, food services, and physical plant. These modelling processes also explore differences associated with the health authority directly delivering the service versus a non-profit or for-profit agency providing the service. [We are] starting to look at actually getting more money in that we can get more  168 value for our dollars in this community agency, and then it’s starting to look at a model. Okay then, how would we then do this in other cultural and language groups. So the Punjabi group. Should we be giving money to [the non-profit society] or some other group to develop, what we’re trying to develop in-house. Which we’re not very good at. The preceding speaker articulates that health authorities are not always the best entities to deliver certain kinds of health services. The participant highlights that ethnocultural-specific social service agencies may be better suited to deliver services to specific sub-populations. Decisions often require something to be implemented and participants identified two dynamics that impede follow through. As documented in chapter five, several participants identified that four-year election cycles impact health authority long-term planning and change management initiatives. While existing programs with longstanding budgets may see more stability, pilot projects and health system change initiatives often rely on one to three year budget timeframes with resources coming from inside the health authority or from specific government initiatives. These participants argued that short budget timeframes make it difficult to plan a change, implement it, and examine the impact over a long enough period. In situations where the health authority wishes to change health outcome patterns, it becomes unrealistic to see shifts in outcome patterns for upwards of fifteen years. Also stated in the previous chapter, political interference can stop a decision from being implemented. Participants recounted a variety of experiences where challenges have come from stakeholders who participated in the decision-process, as well as people who react to the announced decision. While participants stated that formal decision-making processes  169 help to build transparency and clarify why the unpopular decision was the right decision from the perspective of the health authority, two participants who highlighted the political nature of how some decisions are made cautioned that formal processes do not guarantee that the decision will be implemented. Look at the mental health team that we’re trying to move. They’re in a substandard building. They cannot be there any more and they found a new building to expand services and move in. And there’s a huge push back from the community and residents. So even though we spent a lot of money developing the site and stuff, we may have to abandon it. The reason the interview participant gave for stakeholder attempts at stopping the development of these services in their area was that they did not want consumers of mental health services in their neighbourhood. Despite the use of formal decision-making procedures, the decision under discussion may not be implemented. In sum, this participant provides a cautionary tale: formal planning processes may assist the health authority in becoming more systematic and transparent in their decision-making processes but complications may remain unavoidable. The story is left unresolved, leaving the reader to contemplate the ongoing political tensions that health authority administrators must navigate well after a decision is made. In this section I have examined situations where participants are more likely to utilize formal or informal decision-making processes. Timelines, shifting resources across programs or areas of the region, stakeholder positions, decision-maker experience with the possible solutions, and the degree of risk were identified as key influencing factors. I ended this  170 section by identifying that short-term funding commitments and competing stakeholder priorities can derail decision implementation. In the next section I present the tools that study participants identified using during formal decision-making processes. 6.3 Decision tools Figure 6.3: Decision tools  All study participants described using a variety of tools during decision-making processes. These tools were mostly discussed in relation to formal processes,46 and participants stated that they use these more frequently since the health authorities grew in size in late 2001. Interview participants argued that the tools help them organize their work processes, as well as strengthen health authority transparency and accountability. In many instances, participants find these procedural tools useful; however, with the increase in the  46 This does not preclude the usefulness of the presented tools with informal processes.  171 use of these tools, comes a reduction in their ability to reflexively shift directions midstream when they deem necessary. Table 6.2 describes a series of tools used by participating administrators in BC health authorities. I examined all transcripts and accompanying documents for any instances where tools were described. I looked at the examples and, based on this, developed a description of each.  172 Table 6.2: Formal decision-making structural tools Tools  Description Business cases  Describes project goals and objectives, alignment with strategic directions, project short and long-term costs and benefits, risk assessment as well as an evaluation model. Builds project clarity. Builds accountability by assigning executive team sponsorship. Often a time intensive process. Risk assessments  Often part of business case but can also stand-alone. Lists all risks associated with the project, the probability that it will materialize, the impact if it materializes, and the proposed mitigation strategies. Example: Health authority needs city permits and to comply with bylaws prior to providing Audiology services in a new building. Administrators identify a low probability and low negative impact. They propose a mitigation solution: early engagement with municipality and affected public. Gantt charts  A bar chart that outlines important project elements. The chart shows the start date, the projected completion date, and the progress made on each element. Can also compare projected and actual progress. Can be used in association with PowerPoint presentations and other communications tools listed below. Committee structures  Examples provided comprised of committees that include only health authority employees, and a mix of health authority employees and other stake-holding agencies. In some cases there may be health authority-appointed community members. Each committee recognized by the health authority develops a terms of reference document. Terms of reference  Used in association with planning committees. Describes the reason for committee, accountabilities, committee membership, committee functions, and meeting schedule. Stakeholder engagement (may dovetail with committee structure)  Planning processes may develop stakeholder engagement plans. The plans include identifying who is considered a stakeholder, what kind of information they require from the identified group(s), and how they will gather this information. May also include identifying the level of engagement from informing at one end of the spectrum to involving key  173 stakeholders in the decision-making process at the other end of the spectrum. Modelling options  A process of collecting all relevant information and examining what will happen to an existing or proposed program when staffing hours or types of staff and other variables are altered. This assists the administrator to more closely examine their options prior to agreeing to make a large change. Proposal scoring templates  Health authority establishes proposal priorities. Each submitted proposal is weighted against these priorities. Scores are tallied across the review committee. The highest scoring proposals are often forwarded for funding consideration. Communication documents: • Briefing notes • Question & answer documents • Newsletters • Update reports • PowerPoint presentations • Press releases Documents used internally or externally. The messages are tailored to the audience and serve different purposes. Internally the documents are vehicles for sharing information, to align management messaging about a specific topic, and to maintain an issue a priority at a certain level. Externally the documents help keep people informed, shape public perspectives in a specific direction, sometimes draw attention away from other contentious issues, and can provide opportunities for more active engagement through newsletter surveys. Evaluation plans  Outcome measurement frameworks establish the funded service or initiative’s purpose, expected long term and short term outcomes. Reporting mechanisms are also established. As previously stated, the tools described in Table 6.2 are drawn from participant stories about how they make health authority planning decisions as well as participant- supplied documents. Participants consistently spoke about developing business cases and performing risk assessments using committee structures, developing terms of reference documents, and using communication tools. In contrast, I found far fewer situations where participants described using technically involved modelling techniques that are often used in  174 decision-analysis decision-making processes, nor building formal engagement strategies. When participants described proposal scoring processes, scoring occurred in relation to two specific areas: internal high-level program and project-specific priority setting exercises; and to evaluate requests for funding from external agencies gathered through formal call-for- proposal processes. Each time scoring processes were described, participants also indicated using a mechanism where stakeholders could appeal the final decision — a consideration absent in all other types of decision-making descriptions. 6.4 Evaluating decisions Figure 6.4: Evaluation  Decision tools alone do not guarantee a good decision. I asked all interview participants what tells them that they made a good decision. In response I received a variety of answers from “I just know” to more involved descriptions of the importance of both procedural and substantive considerations. In this section, I examine what makes a decision  175 “good”. These good decision indicators relate to how the person came to the decisions, gut feelings, and whether anticipated decision outcomes were achieved over time. In relation to immediate indicators and similar to identifying the reason for promoting the use of tailored health care services and initiatives, participants stated that they knew that they had made a good decision because intuitively they knew “it [was] the right thing to do”. Another participant describes a multiple stakeholder decision-making process. One can gather that information and document what can be documented. And that other intuitive piece is sort of a deeper knowing that is much more difficult to document. I think it’s more acknowledged and I have seen it. I have experienced that acknowledgment around a table and it comes in appreciations. I’m not sure how else to describe that. This participant argues that a decision is good when multiple stakeholders are supportive of the planning direction. S/he argues that not all indications of whether a decision is good come in tangible forms. The indicators also come in the form of good feelings expressed or collectively experienced. These intangible indicators appear to act as useful heuristics but do not appear to be as useful in larger health authority formal documenting processes. At most the formal application of collective appreciation may take the form of a set of meeting minutes indicating widespread committee agreement for a planning decision. While the actual decision-making process could be evaluated immediately, it appears more difficult to establish whether the resulting decision was good. In such instances, participants argued, time must pass in order to gain a greater sense of the results. Participants described using formal and anecdotal evaluations as a way to establish whether an  176 implemented decision was effective and achieved intended outcomes. Proxy measures are often used for both immediate and longer-term indications that the decision was good. In these situations, participants identified receiving information through staff feedback, radio/television/print news, elected officials and public feedback. Interview participants argued that if no one complained about the decision, then this helps them know that the decision was good. However, this silence can be problematic. As one participant argues, these are the decisions that are sometime most troubling: I think sometimes poor decisions are made and you’re neutral. You don’t go backwards or forwards. And maybe we don’t pay enough attention to that. I do think that happens actually quite a lot. Where there’s a relative apathy about what we do. Or the decisions that we make just because something’s not giving us trouble and we have enough trouble somewhere else. We’re good, leave it alone instead of paying attention to it … Or feeling, “Oh well, no news is good news. Let’s carry on”. As such, this administrator argues that s/he needs several pieces of information in order to know whether a wise decision was made. However given the fast paced environment health care administrators work within, evaluating decision outcomes may not be as high of a priority. I have identified three services-specific considerations that assist participants in making what they would consider a good decision. First, any proposed change must ensure patient safety. Second attempts must be made to ensure that services are not compromised. This could come in the form of no service reduction or an enhancement in service through  177 restructuring processes. Sometimes services must be reduced and therefore participants also examine who else can provide the services no longer provided by the health authority. Participants also indicated the importance of building transition plans for patients or clients that use services that are being eliminated. Third, the participant is certain that the affected staff have the capacity to deliver the services in the way and volume required by the new service structure. These considerations are articulated in the following participant quote: [You will have] achieved what you thought you’d achieve and you haven’t undermined the service. And you haven’t put patients at risk and you haven’t stressed the staff beyond their coping abilities … that you’re still meeting the need that’s out there, albeit in a different way. While decision-making process and actual decision outcomes could be evaluated, three additional aspects related to health system decision evaluations remain difficult for participating administrators. First, one participant spoke of the difficulty of effectively evaluating the cascading impact of decisions made in one part health authority system on another area. We talked about [the need for] some kind of decision-making tool to help us make the decisions that we have to make. Sometimes you have to make a decision very quickly and you can’t go through the long process. But what we wanted to do is have a decision-making framework … We have to look at the whole system, and I think if we’re saving money, we have to really save the money. It can’t be on the back of another service or another department. For instance, decisions to eliminate mental health inpatient beds in an acute facility may increase the demand on community-based housing and counselling services. Without  178 strategies and resources directed and mitigating potentially negative effects of the original decision, new problems may arise. Although participants may consider this fairly predictable outcome, they identify that other unknowable unintended impacts are less easy to attend to until it is often too late. Next, participants recounted difficulties related to tracking or evaluating the impact of providing particular services that are short-term in nature. While it may be possible to demonstrate how providing foot examinations for diabetics consistently over a period of years can reduce the likelihood of foot ulcers leading to amputation, it is much more difficult to track clients who use a short term service and to follow them over an extended period of time to see if it reduced their emergency room usage associated with their particular diagnosis. A participant provides a similar example and sees potential long term hope for addressing this current concern. What we mostly have is volume, utilization data, which only tells a little piece of the story. We don’t have a lot of qualitative data. And we don’t have a lot of long-term data. Like looking at children over time. What happens to them five years, ten years out? We mainly have the number of kids served, the number of visits, and number of clients … I don’t know if we had better longitudinal qualitative client outcome based data [whether] it would help us with our decision-making. But often we have to make decisions quickly and all we have is numbers now. Hopefully that will change with [the introduction of a health authority information system] and to be able to actually look at, outcomes over time. Currently, however, rules associated with how health authorities are allowed to use existing  179 patient data for health system improvements combined with multiple incompatible client file databases within and across health authorities make evaluation difficult to realize. Due to the interview question construction, most participants provided information on how they evaluate decisions that they make. One participant, however, highlighted his/her concern about the impact of refusing to choose to fund an initiative. Most of the things [that] we think back on in our career [is] that out of our good decisions are things that worked out really well. But we make probably more good decisions where we said “no”. But you know, what do you measure then, right? And it may have been a good decision to say no. But you can’t then point to a stream of data afterwards and say “Well, I was right, I should have said no, and I did say no”. While learning from successful experiences where a participant said “no” to a proposed issue or option may assist the participant in building decision-making expertise, there may be limited ability for health authorities to pursue this avenue of evaluation given their currently minimal access to evaluation expertise. 6.5 Conclusion Participating BC health authority administrators understand that their decisions are value driven. A significant degree of similarity spans the interview dataset when it comes to clarifying the series of principles that guide participant decision-making processes. Fairness, appropriateness, and collaboration were important considerations that participants raised in relation to their decision-making activities. As situations requiring participants to make  180 funding decisions are identified, so too are the first of many tradeoffs that they must make, namely, whether to use formal or informal decision processes. Formal processes have become more frequently used in recent years. While participants hailed this as a step in the right direction, they remained reluctant to reduce expert judgement to a series of procedural steps. Participants identified situations where formal and informal processes are more likely to be used. I placed five participant-identified factors on a continuum to graphically show the shifting nature of contextual elements that shape both formal and informal decision-making actions. Factors included temporal, resource distribution characteristics, stakeholder, experience with proposed options, and risk considerations. Next I presented participant descriptions of a series of rhetorical tools that they identify making use of during decision-making processes and to communicate their results. Not all described tools are used in every formal decision-making process and some are used during informal processes. When the tools are used during formal decision-making processes, they assist participants in establishing a decision-trail that can indicate what the issue under consideration was, who was involved in the process, and how decisions were made. It is hoped that fair-minded people will agree with their proposed direction. Participants provided information on how decisions are made and how the results of the decisions are evaluated. For the most part, evaluating the procedural elements of the decision-making process can be made immediately. Participants who use both formal and informal decision processes are able to point to key information that they used to lead them to their decision choices. Participants from all four health authorities stated that they “just  181 knew it was the right thing to do” or that the choice was “common sense”; however, using formal decision tools help to tell a more detailed story about how they came to understand this common sense position. The degree of stakeholder satisfaction or reaction to the decision process also provided participants with important information when evaluating how successful their process was. Following formal decision-making procedures does not guarantee that decisions will be implemented. While participants can produce a decision trail that will provide a narrative about how a good decision was made, even fair-minded stake-holding groups who are not happy with a chosen direction can employ organizing strategies to block the health authority from moving in their chosen direction. Dissatisfied stakeholders may engage in social action using unions and professional associations along with media and elected officials to build support for their concerns. As stated in previous results chapters, throughout this process, health authority administrators must respond in ways that help to position and reposition their decision choices. Sometimes participants are successful in implementing their original decisions, at other times they must alter their decisions in small ways, and occasionally they must abandon their decisions completely.  Beyond evaluating the procedural elements of the decision-making process, participants stated that they must often allow time to pass before they can be certain that they produced good decisions. As time passes, participants either formally or informally assembled proxy measures that tell them that the system and their employees have the capacity to work within the newly designed environment, that indicate that intended short  182 and long-term outcomes have been achieved, and that stakeholders are satisfied or at least relatively silent about the changes. Silence, however, can also lull participants out of recognizing and addressing a poor decision outcome since other pressing concerns consistently outweigh them in priority. In chapter seven I expand on how decision-making styles and strategies outlined in the current chapter align with and diverge from existing published decision models. I also examine in greater detail the issue of how competing stakeholder group definitions of what makes a decision good can complicate decision implementation, and I apply deliberative democratic literature to this discussion.  183 7 DISCUSSION 7.1 Introduction In this chapter I reflect on my research results, apply these results to the literature, and come to a series of conclusions. I have organized this chapter into seven sections. (1) First, I discuss my research results on participants’ views of what makes a good health care system. (2) I then discuss their views on issues related to the use of tailored health care services. (3) Next, I explore my results associated with decision-making styles. (4) I then move to a discussion of the role of public involvement in health authority decision processes. (5) Issues related to relations of power were raised throughout the participant interviews. As such I provide an analysis of power relations significantly influenced by Flyvbjerg (2001). (6) In the sixth section, I provide a series of research limitations, and (7) in the final section I conclude. Throughout this chapter, I test my three original research assumptions originally introduced on page twenty-nine. 7.2 Good health care system Significant agreement exists amongst the interview participants when it comes to perspectives on what makes a good health care system. During these descriptions I found that participants support the Canada Health Act criteria and by extension the Medicare program. In addition, participants provided concrete examples that align with the Romanow Commission report recommendations. These Romanow Commission recommendations were  184 arrived at after an extensive 18-month federal government-led process that engaged experts and tens of thousands of citizens, in addition to reviewing empirical evidence (Romanow, 2002). In chapter four, I provided study results that parallel several Romanow report recommendations. These included increased attention to health care system governance issues such as transparency, and accountability;47 shifting priorities from acute care towards a primary health care focus; and the increased attention to mental health, elder care, and capacity to self-advocate. Aligned with these proposed health system changes, participants proposed that the health care system should be conceptualized as a wellness rather than illness focused system. I also found that participants supported the social contract belief that we have a collective societal responsibility within a public health care system to care for those with poorer health outcomes. This manifested in statements such as “it’s the right thing to do”, “we can’t afford not to do it”, as well being echoed through supporting documents related to structural commitments and multi-governmental agreements that address Aboriginal health (First Nations Leadership Council, 2007) and core functions in public health (Population Health and Wellness, 2005a). It is clear that participating administrators located within BC health authorities understand that there is a direct health authority responsibility to address health differences and inequities across the population they serve. Thus, I have observed a  47 Transparency and accountability considerations will be addressed below in relation to decision-making.  185 coherent alignment between participants’ stated beliefs and organizational and governmental requirements associated with collective duties owed to members of the public within their geographic boundaries. While participants demonstrated an understanding of the importance of social determinants or non-medical determinants of health (NMDH) and their effect on individuals, they provided very few examples of attending to them in any coordinated or comprehensive fashion. Most often, participants provided stories about how they went about reducing or eliminating these resources in times of deficit or re-investment. In a recent national study of health authorities on their involvement in addressing the determinants of health, Frankish et al. (2007) found a similar trend: determinants of health more closely associated with clinical practices such as addressing healthy childhood development and personal health were much more likely to be resourced, whereas issues related to income, social status, employment, and working conditions had limited long-term programming or core funding associated with them. In chapter two, I made three assumptions. In the first assumption I suggested that when it comes to funding, reducing funding, or eliminating funding to targeted health services and initiatives, health care administrators draw on implicit ethical principles framed by the Canada Health Act (Health Canada, 1986), and implicitly mirror a social contract perspective (Mendelsohn, 2002), based on the belief that the collective has a moral responsibility to support individuals impacted by unequal social relations (race, class, gender, etc.). I found this assumption held up. However, participants demonstrated that they also use  186 a series of procedural or governance-related principles that include: transparency, accountability, appropriateness, effectiveness, and collaboration. These principles complement the Romanow Commission recommendations as described above. Initially raised in chapter five and expanded in greater detail below, these additional considerations make it difficult to pursue certain equitable practices given the requirement to attend to the needs of the entire population. 7.3 Beliefs about tailored health care services In chapter five, I ordered a broad range of study participant responses related to the value of using tailored health care services to address the unique needs of sub-populations by placing them on a strategies for change continuum. It is obvious that all participating health authorities use tailored services to address access issues, but it is less clear that participants believe that this strategy will lead to closing the health differences gap. I propose that a likely reason for this is that sub-population needs may be initially framed as issues of difference or inequities in health outcomes; however, when participating health administrators wish to address these health differences, they find that there are currently too few options to address this goal. As a result, participants gravitate towards an existing strategy of developing tailored health care services that unintentionally reframes the initial problem as an issue of access. In short, what starts out as an attempt to address health differences or inequities quickly reduces to issues of access to specific health care services — an area that health authorities have a majority of control over. Further, I have found no examples where  187 participating health authorities are evaluating the long-term health outcomes associated with the populations accessing tailored health services. This may be the result of it being easier to evaluate access issues by gathering utilization statistics than it is to evaluate health outcome changes, and that so many other contextual arrangements, such as where a person lives and what s/he eats, influences a population’s health. As highlighted in chapter six, one health administrator admits: What we mostly have is volume, utilization data, which only tells a little piece of the story. We don’t have a lot of qualitative data. And we don’t have a lot of long-term data. Like looking at children over time. What happens to them five years, ten years out? We mainly have the number of kids served, the number of visits, and number of clients. I propose that participants currently have few alternative ways to attend to this obvious gap in access and its causal link to changing health outcomes. If other options were to surface I propose that more health administrators would contemplate using them as an alterative to relying on tailored health care services alone. Because participating health authorities are able to show that the tailored health care services that they have developed address issues of access but do not show any observable connection to ultimate changes in population health outcomes for these targeted populations, I suggest that health authorities ought to complement their current efforts by building more and stronger intersectoral partnerships. It is widely acknowledged by the study participants and in healthy public policy literature (Fafard, 2008) that much about health (e.g. physical activity, eating habits, education, and employment) is mostly located outside of health care  188 system control. While within my dataset and associated literature (Frankish et al., 2007) I have found some examples where administrators pursue collaboration and partnerships with community and other sectors in order to address so-called wicked problems, it is clear that there is room to increase health authority focus on this strategy. In BC, opportunities through existing and recent health reform announcements provide an excellent opportunity to strengthen health authority and non-profit social service society partnerships. For example, the province recently funded health authorities to develop “Integrated Health Networks” (IHN) — an organizing concept that arranges General Practitioners, health professionals, and non-profit organizations within health authority geographic boundaries to deliver coordinated care for select populations (British Columbia Ministry of Health, 2007a).48 Among other things, this strategy is intended to build stronger alliances between private GPs, health authorities, and non-profit social service societies to attend to patient-related NMDH needs. Additionally, provincial work carried out under the direction of the BC Framework for Core Functions in Public Health (Population Health and Wellness, 2005a) significantly widens the frame of attending to population health from one that addresses the health of individuals exemplified by the Integrated Health Network Strategy, towards broader actions related to health promotion (such as healthy public policy), health protection through legislation and taxes, and health assessment and disease surveillance (Population Health and  48 For further discussion related to organizing care in a new comprehensive way, see BC’s Primary Health Care Charter (British Columbia Ministry of Health, 2007b)  189 Wellness, 2005a; Population Health and Wellness, 2005b). 7.4 Decision-making practices I have found that study participants implicitly and explicitly draw on a series of governance and bioethical principles (See Figure 7.1).49 They recounted stories where they considered what was right, fair, and just on the one hand, while contemplating funding services that would service many people rather than a few, on the other hand. While participants may not always articulate using these principles, they nonetheless demonstrated using them in their narrative accounts about how they make funding decisions. From these explanations, it is clear that participating administrators understand that their decision processes and the procedures they employ are value laden and not simply technical in nature. Figure 7.150 provides a visual representation of two sets of principles used by administrators during funding-related decisions articulated throughout the three results chapters. On the left side of the figure there are four principles traditionally employed through procedural or governance ethics. The right side shows the use of four bioethical principles more commonly associated with health care. These two sets of principles are  49 On page 89 I note the difference between implicitly acknowledging using a specific principle through the use of practical examples versus explicitly naming its use. 50 This diagram is presented as a visual representation only. These principles are not prioritized, nor are they weighted.  190 complemented by a commitment to collaboration across disciplines and sectors. Combined, these principles balance duties owed to individuals and communities while ensuring that public dollars are used wisely. Administrators working within BC health authorities are noticeably aware that their decisions go well beyond who is served and how these services are delivered: they must also attend to public confidence. Figure 7.1: Decision-making influences  Participants provided ample evidence that they employ both formal and informal decision-making styles. In chapter six, I described a series of contextual factors that increased the likelihood that participants would choose one style over the other. Factors included temporal, resource distribution characteristics, stakeholder perspectives, experience  191 with proposed options, and risk considerations. The benefits of understanding the role that these factors play in choosing amongst decision style framing are threefold: (1) the factors can act as heuristic markers for accelerating decision-style selection; (2) administrators can be more explicit about why they chose their decision-style tailored to the specific issue at hand; and (3) less experienced administrators can step up their learning around how to quickly arrive at a decision-style strategy. Patrick Fafard (2008) describes an observable shift in how policy analysis is conceptualized. He argues that there are numerous examples where governments have made policy decisions that rely on more than empirical evidence related to internal policy objectives. Additional considerations include public opinion, election cycles, risk management, and decision-maker preferences. As such, he argues that policy processes are not evidence-based but rather evidence-informed. I have found this same arrangement in my current research. In chapter six, I demonstrated that participating administrators draw on a host of contextual considerations that go well beyond examining empirical or technical evidence during both formal and informal decision-making processes. As administrators either implicitly or explicitly establish their choice of decision-style, they also begin to articulate how they will frame the problem, identify contextual considerations, choose who will be allowed to participate in the decision-making process, and ascertain what will count as evidence. During this process, participants identified considering more than just ethical principles while making decisions. This mirrors and confirms Beauchamp and Childress’ (2001) assertion that government or policy judgements are not guided by moral principles  192 alone. Rather, they argue that decision-makers consider the “symbolic value of the law”, cost of the program and its enforcement once implemented, and demands of competing programs or tradeoffs (ibid., p. 9). Study participants have confirmed this understanding by consistently arguing that while serving a particular sub-population is a prima facie right, in certain situations it is not always the right thing to do. Rather, they suggest that the need must also be weighed in relation to all other health system needs as well as other key contextual considerations. Participants provided me with a variety of case examples where they developed and used rational and procedurally fair ways of making large organizational decisions. These range from using project management frameworks that include developing project charters as well as engagement and communication plans, developing proposal scoring templates to compare funding options, allowing in some cases the opportunity for stakeholders to formally request decision revisions, and building normative frameworks that guide administrators in situations where funding cuts must be made. However, the examples that participants provided do not neatly map onto the existing procedural and normative decision- making models outlined in chapter three. For instance, decision-analysis provides a step-wise procedure that helps decision-makers work through complex problems (Clemen & Reilly, 2001; Keeney, 1992). While participants described a logical and formal process for bringing mostly internal health authority stakeholders51 together to define the decision situation,  51 Participants usually identified that stakeholders who were non-health authority employees were engaged after  193 understand the objectives, and examine potential options, they provided no examples of explicit value-weighting of individual stakeholder preferences in relation to the problem elements, nor to the selection of decision inputs. Without this step, it is very difficult to model alternate decision outcomes for stakeholders by showing them what would happen if they shifted their personal preferences on a specific decision element in relation to the entire decision situation. For example, an individual participating in the decision selection process may value the importance of serving many people over the ability to demonstrate significant cost recovery associated with fewer people from this specific population requiring emergency department services. By modeling the scores associated with each evaluation criterion, an overall calculation may shift significantly if the individual can visually see that altering his/her priorities slightly and in relation to all other participating evaluators in the current decision frame could result in an entirely different chosen direction. The alternate direction may still achieve his/her personal and collective goals but in a slightly different way. Furthermore, the participants did not provide examples where they indicate using any decision analysis tools such as fundamental hierarchies, means-ends objectives networks, or decision trees. As a new addition to their priority-setting toolkit, several participants did acknowledge using the Programme Budgeting and Marginal Analysis (PBMA) framework.52  the decision situation and key objectives were established. 52 For a full description of Programme Budgeting and Marginal Analysis, please refer to Mitton and Donaldson (2004).  194 This tool helps administrators apply economics-based methods to establishing priorities, and identifying areas to reduce funding in order to reinvest in high priority areas. Similar to decision-analysis, PBMA attempts to make explicit participating decision-maker values, engage stakeholders, and leave a transparent decision-trail (Mitton et al., 2003; Mitton et al., 2006; Mitton & Donaldson, 2002). Through this process, administrators make explicit the tradeoffs associated with selecting one choice over another. I have identified two streams of formal decision-making within participating health authorities. The first stream relates to developing formal calls for proposals and the second relates to large-scale system redesigns. A different approach to choosing between options exists. In the first instance, regardless of whether the health authority is reviewing requests for funding from non-health authority operated health and social serving agencies or from departments within the health authority, there are far more procedural tools used in efforts to shape how the selection committee decides between proposals. These include placing all proposals side-by-side and evaluating them with a consistent scoring template. These sorts of practices can be observed in situations where PBMA processes are undertaken, where new resources are being provided by the Ministry of Health, and where the health authority has established strategic priorities and is prepared to invest in select areas.53  53 I found examples where funding strategic priorities by scoring proposals existed in health authorities that are currently using PBMA and in those organizations that are not using PBMA.  195 In situations where health system redesign efforts were highlighted and excluding situations where PBMA procedures were used, I found no data that would suggest that administrators who employ formal decision procedures compare options in a side-by-side manner. This suggests that, in the main, when health authorities apply formal decision procedures, they quickly reduce the scope of their work to the most plausible available solution rather than developing several preferences, scoring each against a set of criteria and selecting the highest scored option. This heuristic approach to narrowing the solution scope is reminiscent of Gary Klein’s Recognition-Primed Decision Model (RPDM) (Klein, 2003; Klein, 1998) more often associated with individual informal decision-making processes. I will expand this discussion below in relation to informal decision-making processes. One of the greatest benefits observed in formal decision-making processes articulated by health authority administrators in each participating health region was their ability to produce a robust decision trail in efforts to show how they came to their decision. This characteristic of a procedurally fair practice was achieved through the use of a variety of tools that included terms of reference, project charters, meeting minutes, and communications documents to demonstrate what the issues were and how they arrived at their decision. In situations where participants spoke of using informal decision processes, their focus was on finding a relatively quick solution that was “good enough” rather than ensuring that a decision-trail was produced. This is where I found the highest number of examples associated with participants acknowledging that they employ heuristics to help them quickly  196 arrive at conclusions. For example, one participant described sizing up people who were providing him/her with the additional information that s/he needed to make his/her decision. The participant used the representativeness heuristic (Bazerman, 2002) to assess what s/he knew about the person’s analytical skills and how useful his/her feedback was in previous situations. In this situation, past experience with a potential decision support person informs the current decision-situation. This participant used a similar strategy when s/he evaluated the truth-value of existing data sources. A different participant argued that s/he feels more confident when community leaders from organizations with a good reputation in their communities provide guidance on what should be done in a particular situation. A third participant demonstrated his/her use of the anchoring and adjustment heuristic (ibid., 2002) when s/he described his/her ability to use past experiences to quickly calculate the likely effect that a proposed cost savings strategy would have on a variety of future states. For instance, if a program is completely closed down and one year later the Ministry of Health wants to fund that kind of service, the health authority will have less opportunity to attract those new dollars to their communities because they no longer have the infrastructure necessary to quickly start up the service and show results. Fifteen of the twenty-four participants crossing all participating health regions and representing both rural and urban experiences demonstrated a strong commitment to using formal decision-making procedures in situations where the conditions allow. Even within this segment of participants, it is still evident that they employ what Klein calls informed intuitive practices (Klein, 2003; Klein, 1998). For instance, one administrator identifies what could be  197 considered thin slicing (Gladwell, 2006). The participant describes quickly scanning or assessing existing facts and using them to shape the decision situation. After the decision situation is framed, s/he then begins to shape the actual decision-making procedure to align more with what could be considered a formal decision-making process. Similarly, a different participant describes using a pattern matching function during his/her formal decision- making process where s/he assesses from past experience whether producing a mainstream service that integrated the needs of both mental health and addiction issues would make more sense clinically and for the target population, or to create a new specialized service apart from the existing mental health program and the existing addictions services program. In this example and consistent with Klein’s RPDM, the participant appears to draw upon his/her existing mental models (similar past experience), and then applies a mental simulation (by asking “how well will it work in this present situation”) to evaluate whether it is worthwhile expanding the examination of the plausible workable option. Participants also described looking for cues that help them quickly assess if things are going well or whether there are problems that need to be addressed. In chapter six, I described a situation where a participant wanted to ensure that his/her team was acting in a collaborative way. If s/he sensed that team members were “defensive”, it provided him/her with a trigger that more attention ought to be focused on creating the conditions for a collaborative environment. This situation provides another good example of how cues are used as a heuristic marker for the decision-maker. In this case the participant is able to highlight that s/he is consciously scanning his/her environment to pick up on indicators that  198 his/her strategies are working or that s/he requires additional attention in order to correct his/her course.  One participant lamented about what s/he perceived to be the loss of decision latitude, or the degree to which an individual has personal control over his/her workplace decisions (Ganster, Fox, & Dwyer, 2001), given increasing expectation that s/he use formal decision-making processes within his/her organization. Hubert and Stuart Dreyfus’ theory of human learning is quoted in Bent Flyvbjerg (2001). In this theory, there are five levels of human learning. Flyvberg’s summary is listed verbatim below: (1) Novices act on the basis of context-independent elements and rules. (2) Advanced beginners also use situational elements, which they have learned to identify and interpret on the basis of their own experience from similar situations. (3) Competent performers are characterized by the involved choice of goals and plans as a basis for their actions. Goals and plans are used to structure and store masses of both context-dependent and context-independent information. (4) Proficient performers identify problems, goals, and plans intuitively from their own experientially based perspective. Intuitive choice is checked by analytical evaluation prior to action. (5) Finally, experts’ behavior is intuitive, holistic, and synchronic, understood in the way that a given situation releases a picture of problem, goal, plan, decision, and action in one instant and with no division into phases. This is the level of true human expertise. Experts are characterized by a flowing,  199 effortless performance, unhindered by analytical deliberation. (Flyvberg, 2001, p. 20-1)54 In light of Dreyfus and Dreyfus’ theory, the participant’s concerns related to the loss of decision latitude can be seen as an important outlier perspective in my dataset. It stands in contrast with other highly experienced administrators who state that they value using formal processes to establish procedurally just conditions and a higher likelihood that decisions will be implemented, as well as with other less experienced participants who state that they are thankful for more formal processes because it allows them to share the responsibility and criticisms for decisions that were made. The outlier in this instance brings attention to what could be lost due to a system becoming too rigid in its practices. Thomas Homer-Dixon, speaking at the Canadian Public Health Association Annual Conference in 2008, argued that the Canadian health system has become so well adapted to a set of circumstances in response to growing complexity by becoming efficient and productive that when a crisis develops the system cannot attend to it. I suggest that, through the drive to become a well-oiled machine, health authorities are erasing the decision latitude that expert administrators require to keep the health system from becoming too rigid in the face of rapidly changing needs. A balance must be struck whereby the public and professional associations have confidence in the tough decisions being made on their behalf (a call to rationality and decision-trails that fair minded  54 This theory does not suggest that as people move through the stages that they become increasingly co-opted by the system, since we retain the ability to call our actions into question and make changes accordingly.  200 people can agree to), while allowing increased appreciation for informed intuitive and flexible informal decision processes. I have found that two of the four health authorities that participated in my research have established normative decision-making processes, especially in situations where funding cuts are required.55 However, it appears that these models are minimally used. For instance, one health authority lays out a plan for procedures that must be attended to in situations where cutbacks must occur.56 The model prioritizes where the administrator must first look to find savings. In this model, a reduction in the actual amount of service to a population holds the highest value and therefore is the last place that cost recovery is to be sought out. In the normative models that I examined, I found only limited attention to the accountability for reasonableness formula that Norman Daniels and James Sabin articulate (Daniels & Sabin, 2002). In their five-step formula57 that establishes whether the health care rationing process could be considered legitimate, there must be a step whereby the proposed decision can be appealed and a mechanism to ensure that this criteria is upheld. In my study, I found that these criteria were more likely to be present in situations where proposals for funding were evaluated and scored by a committee. In all other situations, participants  55 A third health authority reported being in the process of developing a normative decision-making model. 56 BC health authorities are technically required to ensure that their budgets are balanced annually and therefore cannot carry a deficit. 57 Refer to pages 56-57 of this dissertation for a detailed description of Daniels and Sabin’s four-step process.  201 remained silent on how a decision could be appealed. However, based on study interviews, it is clear that participants understand that if an individual or group was not satisfied with a given decision, recourse was taken through public action. Actions included contacting the media, challenging the decision at local health authority public Board of Directors meetings, and by contacting elected officials. The second assumption I proposed in chapter two was that clinical efficacy and economic efficiency drives decision-making with little attention to ethical and social value. I found that this process was more involved than I anticipated. For example, I found no indications that study participants evaluated the overall efficacy of the programs they described, nor that they could establish whether the tailored programs made a difference in the overall health outcomes of sub-populations. Political activities such as elected representatives calling into question health authority decisions on behalf of their constituents, television and print news articles, and administrator personal preferences appear to influence the decision-making process as much as efficacy and efficiency drivers. This finding aligns with Terry (2004) who identified that finite resources and a desire for re-election have a significant effect on how decisions have been made in universal health care systems. With respect to my third assumption, I stated that formal decision-making procedures are seldom used in health authority decision-making processes unless it is related to a large- scale health system redesign. As indicated in chapter six and illustrated in Table 6.1, I found that participants are most likely to use formal processes where one or more of the following five factors are present: (1) the time frame allows for a longer decision-making process to  202 occur; (2) resources will be moved from one program area or part of the region to another; (3) key stakeholders disagree about how to resolve the issue at hand; (4) the decision-makers are unfamiliar with the proposed solution; and (5) there is a significant degree of organizational risk associated with the problem and its potential solution. I also found that participants with limited administrative experience were more likely to use formal decision-making processes for smaller funding-related decisions. In these instances, I found that the participants described being uncertain about what they should do, and they believed that involving more people in the process would produce a better decision, allow them to leave a decision-trail to account for their ultimate decisions, and allow them to share the blame for unpopular decisions. However, my findings also highlight that regardless of how formal or procedurally fair a decision-making process may be, there is no guarantee that effected individuals and groups will accept the results. 7.5 Public involvement Participants mirrored popular sentiment and academic literature (Abelson & Gauvin, 2006; Chafe et al., 2007; Martin, Abelson, & Singer, 2002) by stating the importance of public engagement in health care system decision-making processes, but their case examples seldom showed significant levels of actual stakeholder involvement. Where descriptions of engagement were highlighted during interviews, I found that participants provided minimal information related to using sustained engagement strategies. For the most part, examples highlighted that they used occasional meetings, focus groups, or surveys. In relation to the  203 proposed continuum of engagement described in chapter three, most examples related to “inform” or “consult” levels and much less to the “engage” level (Rowe & Frewer, 2005). There were a few examples of more involved engagement strategies, but this was not evident across health regions. In those cases, stakeholders participated on funding proposal selection committees where the decisions would affect their community.58 Participant discussions of engagement, however, related to the development of sub-population tailored services for the most part. A review of participating health authority websites provides more of an indication that a variety of engagement strategies have been utilized. These include engaging members of the public in long-term advisory committees, seeking their advice on the best way to build a broader engagement strategy, and empowering an advisory committee with the mandate to evaluate in some form how well the health authority is carrying out its commitments on a particular issue. Again, the types of engagement strategies employed and the consistency with which they are deployed within and across participating health authorities is at best uneven. 7.6 Power Health authorities obtain their right to govern the design and delivery of health care services for people residing within their geographic boundaries through both formal and  58 I am using the term “community” to represent geographic and identity-based dimensions. Examples of identity-based communities include seniors, women, and Aboriginal, among others.  204 informal sources: the provincial government provides health authorities with legitimate power by passing health planning and service delivery devolution legislation (Government of British Columbia, 1996); acquired or procured management rights to reorganize or reform services from local providers and institutions, and from the public who collectively grant the mandate to have their needs, wants, and preferences represented by the health authorities (Lomas, 1997; Lomas, Woods, & Veenstra, 1997). Locally, power relations operate through the hierarchical structuring of employment titles and an administrator’s ability to control agendas and resources. This works through delegating authority (constrained by policy) to the individual. Through this process administrators are able to shape and/or limit what is possible. For instance, a group of administrators may agree that the health authority will no longer allow contracted agencies to use funds provided by the health authority to be used to operate pre-employment programs. These types of program attempt to assist marginalized groups to gain the skills necessary to enter the work force. Employment and access to adequate wages is a NMDH (Health Canada, 1986). Administrators are also able to reduce the future likelihood that their health authority will seriously consider supporting other services or initiatives that address NMDH, citing the importance of focusing limited health service resources on issues more closely associated with changing health outcomes. For example, administrators may support a service for older adults who have co-morbid conditions in efforts to directly reduce the number of times that this population group goes to the emergency department for things that could have been avoided with proper coordinated medical care. The health authority in this case becomes  205 more explicit about where it prioritizes its funding and reduces its commitment to addressing further upstream prevention and health promotion issues. Power relations are also shaped by the way that the needs for tailored health services arrive at, are considered, and/or are ignored by health authority leadership. As described in chapter five, the need for tailored health services can be identified from inside and outside of a health authority. Well-positioned health authority administrators are able to promote service needs informally by speaking with coworkers and building support and formally by developing business cases and communications documents to help position the need in relation to other competing issues. I found that participating administrators return to these tools, in addition to government inquiries and public engagement activity reports, in times where they wished to reinforce a specific position or when they wanted to show that their actions were transparent and accountable. Interview participants provided multiple references to the development of business cases focused on reducing emergency department visits by having primary health care providers build more effective and systematic chronic disease management processes with their patients. This included supporting the General Practitioners (GP) in identifying all of his/her diabetic patients and placing their names on a disease registry. GPs were encouraged to use a flow sheet that identifies current best practice tests that ought to be provided to this disease population group. Next, the GP’s office contacted the identified patients and invited them in for regular visits to ensure that their medical care is attended to in a proactive manner. This development in BC aligns with recent provincial funding arrangements that directly support GPs to change their practice and to align more  206 closely with health authority planning and service delivery (British Columbia Medical Association, 2006; British Columbia Ministry of Health, 2007a). Interview participants also described supporting community groups to develop their business cases or in understanding the best way to frame their issue in ways that align with health authority priorities and in a language that health authority administrators will support. As such, both health care administrators and other stakeholder groups are able to use similar rhetorical tools in their efforts to support or challenge any given health authority direction. Participants identified that stakeholder groups such as professional associations and community groups can place pressure on health authorities through direct verbal requests that decisions be revised or to make decisions that have been set aside. For example, one participant spoke of identifying an inequitable distribution of resources across the region and planned to redistribute resources so that it was more appropriate. The participant stated that the individuals receiving the new resources were happy, but those who were having their existing services reduced or eliminated were not supportive of the changes. An example of stakeholders pressuring the health authority to make a decision that had been set aside includes a situation where access to services that addressed both mental health and addictions issues59 were not being attended to in a way that was accessible to the client population. In both cases, publicly raising awareness of the needs of the target population created the  59 This was described as “dual diagnosis”.  207 pressure necessary for the health authority to re-examine the issue more closely. These examples demonstrate that, in situations where stakeholders are not supportive of the direction the decision is being made, they may contact public outlets to make their concerns known. Interview participants stated that stakeholders often contact elected members of various levels of government, as well as the media, with their concerns. This strategy is used to raise awareness of the issues in question and in hopes that actions aligned with their perspectives will ensue. Establishing the boundaries of what issues health authorities will seriously consider effectively reduces the likelihood that certain types of needs, such as ones that address NMDH, will be addressed. However, even within this limiting frame, study participants state that they compare the competing needs of multiple sub-population groups and attempt to examine possible ways of positively influencing their health outcomes. While I did find that all participating health authorities are becoming more explicit about addressing tradeoffs associated with choosing amongst options that make it to the planning table for consideration, these experiences are sporadic and are not embedded throughout the health authority structure. In cases where strategies for addressing specific sub-population’s needs have been set aside by the health authority because the proposed solutions are no longer supported,60 it would be very difficult for these needs to be considered in the future unless advocates of those groups frame their issues, data, and  60 Examples include running pre-employment programs and preventative child dental care.  208 proposals in a way that aligns with health authority priorities. This could be a very slow process. Yet given the amount of attention that study participants ascribed to political interference and its ability to reshape a health authority’s proposed direction suggests that a series of well placed media stories could accelerate public and health authority support to attend to an issue. Hypothetically, if several people died in an emergency room in a short period of time due to, among other things, communication-related barriers resulting from the medical staff not understanding the cultural and language inferences related to the emergency visit, and if this were reported to the media, it could create the conditions for social action. It is fair to assume that policies may be altered and new solutions for culture and language translation considered. This example may be further influenced by the availability heuristic that people use to assess the frequency, probability, or likely causes of an event by the degree to which instances or occurrences of that event are readily available in one’s memory. As stated in chapter three, if the event evokes emotions and is vivid, easily imagined, and specific, it will be more available than its opposite (Bazerman, 2002). Interview participants describe using various tools to shape how power flows (Flyvbjerg, 2001). A participant described burying a potentially contentious administrative decision by releasing many public announcements at once in efforts to divert media and public attention in another direction. Participants also frame public announcements in positive ways that highlight how patients and the health system will benefit from the changes. While doing so, however, other undesirable effects of the decision may be occluded. For example, the funding used to address the current plan may have been captured  209 through the reduction of service in another area of the health care system and could reduce the provision of effective care in that area. While the current context paints a picture of some health authority administrators wanting to support population health and health promotion strategies and interventions, the BC government and BC health authorities continue to maintain a strong focus on hospital- based services. When attention is placed upon non-acute or community-based health care services, they must rationalize how these services help to reduce hospital costs and wait times. These arrangements make it difficult for health authorities to fully embrace long-term population health and health promotion strategies that are not likely to demonstrate positive shifts in targeted sub-population health outcomes for many years to come. Although acute care service delivery appears to be a Canadian health care system axiom, it is possible to envision a different set of arrangements. These games of power (Flyvbjerg, 2001) associated with building health authority strategies and assigning resources could play differently if the provincial government charged BC health authorities with the expectation that population health and health promotion services and initiatives were of equal or greater priority than acute services. In this arrangement, a new set of tradeoffs would present. It would require health administrators to explicitly examine the kind and amount of acute services that they are able to provide to the public given their requirement to shift resources to disease prevention and health promotion activities. Going well beyond the need for technical information, there would no doubt be a need to increase public discussion. It would also require a systematic approach to public and stakeholder engagement and the use  210 of consistent mechanisms for involving the public in health authority planning. In summary, I have argued that health region administrators obtain their authority to make decisions through provincial legislation and local health authority policies. Administrators frame priorities in relation to health authority strategic directions. They use rhetorical tools in order to justify their directions, produce decision trails, and establish evidence that their decisions are transparent and legitimate. Other stakeholders can, however, use these same strategies to challenge health authority directions. In addition, they can take their concerns directly to elected officials and media outlets in a way that can be more sensational in nature. While the way in which attention is paid to population health and health promotion looks solid, an alternate game of power can be considered. I have suggested that with stronger provincial government direction, health authorities could begin to aggressively shift BC health system priorities. This proposed shift may open greater opportunities for public involvement in health care system reform. 7.7 Limitations In chapter two I outlined how my chosen research design attended to my overall research goals. I described potential research limitations and ethical considerations. In addition, I recognized how history and context shape the entire research endeavour, including who will participate and how the interview process shapes the direction of the dialogue. I also documented the importance of researcher reflexivity as an important step in understanding the role of the researcher in the shaping the results. Lastly, I documented  211 current literature related to evaluating qualitative research and how my study design and results attend to these evaluative considerations. I now reflect on areas where my chosen research design provided limited results and how a different research design could have produced different results. I have identified two additional important limitations. The first relates to a systematic examination of how participants use heuristics, and the second relates to my limited ability to generalize participant perspectives to all health authority health care administrators. Study participants describe situations where they use heuristics that may influence their decision processes. However, it is unclear whether the use of these heuristics actually produces a negative effect. While I looked for the general use of heuristics in the ways that participants described their experiences, I did not directly focus my research on this area. I believe that a research design focused solely on the use of heuristics in a decision-making process that involves extended periods of participant observation, and multiple interviews with fewer participants, would achieve greater heuristics-related insights. In this section I am particularly interested in exploring issues related to how my research findings may have differed if my current study, which is focused on providing a snapshot in time, were recast as longitudinal research. In my study, each participant described at least one case example. Once mapped onto a timeline it became clear that each example described an issue that took between six months to four or more years to resolve. When I interviewed participants they provided me with details about the case example and reflected upon their associated decision-making processes. While my study provided several insights into the decision-  212 inputs that participants describe using and explanations where formal or informal decision- making processes were likely to be used, it was less successful at providing significant understandings of how heuristics aided decision-making and where they were more likely to be a hindrance than helpful. To address these shortcomings, a longitudinal qualitative research design would provide greater depth. For example, an ethnographic study combining participant observation, one-to-one interviews, and document analysis may produce novel findings (Creswell, 1998). In this case administrators would be followed and/or interviewed over a two-year period. The researcher would observe planning meetings and associated activities. In situations where the researcher suspected that heuristics were used to quickly size up a situation or come to a conclusion, it would be noted. When appropriate, and as closely following the observation as possible, the researcher would meet with the participant, present the observation, and invite the participant to deconstruct the identified situation and provide in-depth detail as to what s/he was thinking and how s/he came to his/her decision. While this study would likely involve less participants than are involved in my current study, I believe the depth that the proposed study provides would create the conditions to unpack the everyday use of heuristics in administrator-specific decision-making processes. Within this proposed study direction there would be many opportunities to participate in informing the analysis and conclusions in a way that my current research was unable to provide. In addition, it would provide opportunities to see how participants change their perspectives and approaches over time — an observation that cannot be achieved with my current research design.  213 Like all research in applied settings, this proposed direction might be limited by very practical considerations. Two likely barriers include a health region’s willingness to participate and a health administrator’s ability to commit to being followed and interviewed over a long period of time. Given that fewer people would be required for this in-depth study, it is possible that these identified barriers could be overcome. The second identified limitation relates to my ability to generalize my research results. All participants in my dataset were responsible for community-based services. Therefore the results — especially those related to a desire to shift resources from the acute system into population health and promotion activities — should be read with caution. It should also be noted that there exists a countrywide desire to support a shift in this direction, as evidenced by the Romanow Commission report (Romanow, 2002). Even though all participants came from a similar work setting, they brought different experiences and perspectives as a result of their different backgrounds (race, class, ability, gender, education and sexuality), life experiences, and organizational department affiliations. The timeframe within which my study is located also informs what could be found. My research occurred during a period of relative organizational stability. If this study were to be reproduced with the same participants during a time where virtually unlimited health service spending were possible, as indicated by several participants in relation to another province where they previously resided, and an election was pending, I may receive significantly different case examples and perspectives on their meaning. This does not reduce the significance of my results but rather once again reminds us that time and context significantly  214 influences research findings regardless of the research design. However, it cautions the reader about how much my analysis and conclusions can be generalized. In keeping with my epistemological and ontological beliefs related to the production of knowledge, I leave it to the reader to assess my research decision trail and analytical explanations, and to assign his/her truth-value to what I have produced. If they so choose, readers may wish to explore where my analysis and conclusions provide points of departure for exploring how it may apply to their own understandings of what makes a good health care system, the value of using tailored health care services, and assessing whether the decisions that they are making are good. 7.8 Conclusion There is strong participant support for the Canada Health Act, the Medicare program and the belief that the collective has a duty to support the needs of marginalized populations. In addition, I have found an overlap between participant stated health system priorities with those identified with the Romanow Commission report. Most notably, there is a desire for the system to be more transparent and accountable for its decisions, a heightened shift away from an acute-focused system to one that is primary health care-focused, as well as increased support for mental health, and greater attention to the needs of older adults and self advocacy services. However, I only found minimal indications to suggest that NMDH are being addressed within the health authority examples that participants provided during interviews and in the documentation that they supplied.  215 I found that tailored health care services and initiatives are being used throughout all participating health authorities; however, it does not appear that these activities are being evaluated to show that they made a difference in the health outcomes of the populations being served. At best, the services are established to demonstrate that the health authority is actively eliminating barriers to access for certain populations. I proposed that this situation is in part due to the limited number of strategies that health authority participating administrators have in their toolkit when attempting to address access and health outcome differences and inequities. There is a general global trend towards making bureaucratic entities more transparent and accountable to their stakeholders (Abelson & Gauvin, 2006; Abelson & Gauvin, 2004b). This is also echoed at the health authority administration level. Participants identified a shift towards more formal decision-making processes as an important step in the direction of gaining public confidence in health care service system planning decisions. They support the belief that a procedurally fair process will produce better decisions, even if those decisions are not popular. Standardized processes that utilize a variety of decision support services and tools help the administrator work through tough tradeoffs, but choices must still be made. By examining virtuoso traits (Flyvbjerg, 2001) and applying them to participant descriptions of how they make difficult decisions, it becomes clear that two types of people promote the use of more formal processes: those who promote increased decision-maker accountability and organizational transparency, and those at the other end of the spectrum who are still developing their decision-making skills. This latter group identifies the value of an organized  216 process that will share the decision-making responsibility and guide them through to a final decision. They are reluctant to make decisions on their own even in situations where it may be more appropriate to use an informal decision-making process.  I claim that study participants understand that their decision-making processes are value driven and that they routinely draw upon a variety of principles that align with governance and bioethics considerations. I also showed that participants intuitively consider when they will produce either a formal or informal decision process. Far from tidy categories, participants describe embedding informal processes within formal ones and vice versa; still, it is possible to show that in a given situation one decision-process is drawn on more significantly than on others. By understanding when to use one strategy over another, I have argued that participants will be able to provide additional value-added information as to how and why they structured their decision process in a specific context. Participants identified having to weigh issues of equity with those of utility. While coming to their position, participants drew upon additional considerations that helped them make tough decisions. Similar to Beauchamp and Childress (2001), participants considered key contextual considerations such as the cost of the program and tradeoffs associated with choosing to address one population’s needs over another’s. Public involvement activities can be found in participant identified case examples and associated documentation, but in a very limited way. When examples are provided, these are mostly associated with “inform” or “consult” levels of the engagement continuum outlined in  217 chapter three. I recommended that increased attention to using meaningful public engagement strategies during formal decision-making processes would better align the participant’s stated support for the value of stakeholder engagement with their decision process actions. I ended this chapter with a discussion of power. Using Flyvbjerg’s (2001) analysis of power model, I argued that administrators frame priorities in relation to health authority strategic directions and use rhetorical tools in order to justify their directions, produce decision trails, and establish evidence that their decisions were transparent and legitimate. Stakeholders who were dissatisfied with the decisions that were made on their behalf could use the same tools. Participants identified that stakeholders’ strategic use of elected officials and media outlets is the major way that concerns are raised and often resolved. I ended my analysis of power by proposing an alternate game of power where the provincial government provides BC health authorities with clear direction to aggressively shift BC health system priorities from a system that is illness-focused to one that is wellness-focused. Finally, I discussed the three research assumptions initially proposed in chapter two. Each assumption proved to be correct in essence but much more involved than I initially expected. First, while participants support drawing on principles established within the Canada Health Act and framed by a social contract, I found that participants also consider a variety of governance and bioethical principles along with a commitment to promote a collaborative approach to decision-making when it comes to resource allocation discussions. Second, political influences generated by elected officials and media groups appear to sway  218 resource allocation decisions more than clinical efficiency and effectiveness combined. Third, choosing to use formal decision processes are guided by at least five observable factors that included temporal, resource distribution characteristics, stakeholder, experience with proposed options, and risk considerations. While I initially assumed that formal processes were only used in large-scale health service redesign, I found that participating administrators with less experience also use formal processes in small-scale decision situations where they are unclear about how to address the issues at hand or are worried about how the stakeholder groups will react to the decision. In the final chapter, I conclude by evaluating my intended research goals and objectives, and I provide recommendations for practice and future research.  219 8 CONCLUSION 8.1 Introduction Health care administrators are responsible for the strategic and operational planning and service delivery in their geographic area.61 Beyond hiring and managing staff, a majority of the participating administrators also manage multimillion-dollar budgets that include providing non-profit and for-profit agencies with contracts to deliver health care services and initiatives. The intricacy of their workplace environment is further complicated by government- imposed reporting requirements; health care professions expectations (voiced informally by individuals and through associations or formally through union bargaining); and unique needs of their geographic area identified through planning and public discourse. Examples of government-imposed requirements include the mandatory provision of core hospital services. In addition, the government requires health authorities to occasionally deliver a specialized service such as youth-directed crystal methamphetamine education or initiatives that promote better coordinated care for people with diabetes. The unique needs of a given area or population are established through planning processes that lean on a large number of sources  61 In the case of the Provincial Health Services Authority, the administrators would focus on provincial-based services rather than geographic-based services.  220 that include: epidemiological, demographic and service utilization data; needs assessments, peer reviewed and grey literature; government and technical documents; financial data; service provision benchmarks; and public dialogue structured through health authority processes or expressed through community action and/or media spotlights. Participating health administrators described working within environments imbued with personal, professional, and organizational values as well as those outlined in the Canada Health Act, and government-lead commissions. These overlapping and sometimes competing value positions provide direction for their actions but also make it very difficult to move their health planning decisions towards action. In efforts to increase the likelihood that their decisions will be implemented, participants describe pursuing personal and organizational goals through negotiated agreements with multiple stakeholders that include many different public perspectives and expectations. Even when they develop formal decision-making processes that engage stakeholders in the planning process in efforts to increase decision- making transparency, legitimacy and accountability, they find that there is no guarantee that a crucial faction of the public will allow the implementation of the proposed changes. In these situations the participants describe developing strategies designed to run interference such as by announcing multiple decisions in anticipation that the most contentious issue will be overlooked by concerned public groups and thus implemented with less difficulty. Similar to popular media and academic research findings, the interview participants anchored their discussion on the seemingly ever-increasing global health care budget attributed to pharmaceutical and technology growth on one hand, and the continual expected  221 increase in the direct delivery of services, while contending with shrinking operational budget lines on the other hand. Participants spoke of increasing infrastructure costs associated with land valuation and construction. As well, participants describe ongoing requests for additional resources coming directly from physicians wanting access to new technologies, the public requesting quicker access to services, and the provincial government requiring new services or targets62 to be provided and achieved within assigned time limits. Occasionally new resources will be attached to the new demands, as was the case with the requirement that BC health regions attend more closely to chronic disease management, and the recent provincial announcement of new dollars for a variety of initiatives directed at reducing emergency room and surgery wait times (British Columbia Medical Association, 2006; British Columbia Ministry of Health, 2008). Lastly, as health authorities have grown, so to has their reliance on formal processes that utilize rhetorical devices. Participants described the development of risk management tools and analysis, large communications departments, and reliance upon briefing notes, business cases, and other rituals of power. In the first chapter I identified four objectives that drove my research: to gain a better understanding of how health authority administrators make sense of their own funding-  62 ‘Targets’ are aggregate output-related measures associated with achieving goals. Examples may include reducing service wait times, and increasing access to community-based dialysis clinics in rural areas.  222 related decision-making processes; to look at participant descriptions of how decisions are made align with existing normative and descriptive decision-making frameworks and add to this area of knowledge; to explore what the research participants think about tailored population-specific services as a way to address the health inequities gap; and to examine how participants perceive themselves to be making “good” decisions related to funding, reducing, or eliminating funding to community-based tailored health care services on behalf of the public. In order to achieve these goals, I interviewed twenty-four administrators in four health regions across BC and analyzed interview transcripts and associated documents using a qualitative descriptive analysis combined with a narrative structural unit analysis. In the next section, I summarize what was learned in relation to my research objectives. Afterwards, I conclude with a series of practice and research-related recommendations. 8.2 Summarized Research Objectives 8.2.1 How participants make sense of their own funding decision-making processes, and how participants come to understand that their decisions were ethically appropriate. Participants describe using a series of governance and bioethical principles that help them frame what issues can be considered and how this should occur. They frame their decision situation in a way that informs them whether they need to use formal or informal processes. In both cases participants collect information that allows others to understand that the decisions they have made on behalf of the public are good ones. Participants are more likely to evaluate the results of decisions than the actual decision-making process, although  223 examples were provided that showed an increasing commitment to procedural justice. Sometimes participants argue that their decisions were good because they make good sense. This sense-making, however, can be influenced by the availability heuristic and therefore should be considered with caution. Additional contextual factors allow the participants to understand that the decision inputs upon which the proposed solution is founded have a high enough truth-value thereby allowing the decision-maker to accept the chosen direction. More often than not, however, time must pass in order for the participant to know that his/her decisions were good. In situations where funding reductions were imminent, organization- approved tools designed to prioritize where cost savings ought to be found were employed. Participants identified that they would like a decision-making framework that went beyond existing cost recovery normative tools, but said that it must be embedded within their daily workflow in order to increase the chances that it will be used. 8.2.2 Gathering participant descriptions of how decisions are made and their alignment with existing normative and descriptive decision-making frameworks; add to this area of knowledge. Participants demonstrate using both formal and informal decision-making processes. I found that Gary Klein’s Recognition-Primed Decision Model (RPDM) accurately reflects the intuitive processes that participants describe using during informal decision-making processes and portions of formal decision-making processes. However, in relation to formal decision processes, I found less alignment with decision-analysis processes forwarded by Clemen (2001), Keeney (1992; Keeney & McDaniels, 1999), and McDaniels (2000a; 2000b).  224 Most notably, participants rarely spoke of considering options in a side-by-side manner, performing highly technical modelling procedures, and performing individual value weighting processes. I also found that Daniels and Sabin’s notion of accountability for reasonableness was rarely followed. In the main, once the health authority made decisions, stakeholders were not guaranteed a formal venue through which decisions could be appealed. Lastly, participants were very aware of decision tradeoffs that they must make on a regular basis but they acknowledge that they do not have a systematic way to account for the impact that a decision in one area of the system may have on another area. Participants identified that this concerns them. 8.2.3 Explore what the research participants think about tailored population-specific services as a way to address health inequities All participants identified the importance of health equity; however, participants had different perspectives on the usefulness of funding tailored population-specific services as a way to address the health inequities gap that exists in the BC population. I found that three strategies for change existed in relation to this topic. Participants provided a continuum of responses that included: (1) the belief that tailored services were a good stopgap measure until the existing health care system could accommodate the targeted population’s needs; (2) the belief that delivering tailored services were the actual solution because the mainstream services will never be able to meet sub-population needs; and lastly (3) funding tailored services were a poor use of limited resources. Participants provided case studies related to addressing the unique needs of marginalized populations. In these examples it became clear  225 that, while participants believe that tailored health care services and initiatives may address issues of access, they are less likely to be evaluated to see if health outcomes were positively affected. 8.3 Practice and research recommendations In this section, I provide four practice-based and three research focused recommendations. Each recommendation focuses on a specific aspect of decision-making within large organizational settings. 8.3.1 Practice recommendation #1 Research participants demonstrated a strong commitment to the importance of universal health care, issues of equity, and alignment with several Romanow Commissions report recommendations. While participants stated a strong belief in the importance of population health promotion and the role of non-medical determinants of health (NMDH) in maintaining a healthy population, they also recognize that most of what influences health is in the purview of other sectors such as education, employment, and housing. This may partially explain why the case examples that participants provided show limited health authority commitment to fund services and initiatives directly related to population health promotion and attending to the NMDH. In addition, the way that political pressure is exerted through professional associations, community groups, provincial government departments, elected officials, and the media sector, can help administrators implement difficult changes  226 quickly, slow it down, or stop it all together. These arrangements appear to influence how participating health authority administrators approach and prioritize funding options as much as clinical efficacy and efficiency considerations. Multiple arrangements currently exist in British Columbia to promote increased intersectoral partnerships that are intended to better the health of BC residents at individual and population levels (British Columbia Ministry of Health, 2007b; Population Health and Wellness, 2005b; First Nations Leadership Council, 2007). Health authorities are a key stakeholder in these actions and are in an excellent position to increase their attention on promoting and supporting non-health focused sectors to address the NMDH — areas known to positively affect population health outcomes. Therefore, in addition to addressing health service access issues through the use of tailored health care services, and considering my research data and associated literature related to complex systems, I recommend that health authorities increase their focus on intersectoral partnerships especially with school systems, the social service sector, the employment sector, as well as with municipalities and Aboriginal leaders. The pursuit of intersectoral partnerships would reshape other participating sectors sense of responsibility in relation to the health of the population and potentially leverage their commitment to addressing the identified health-related needs. As representatives come to understand their sector’s role in addressing particular health issues, it may make it more difficult for each participating group to shrug off its associated responsibilities.  227 8.3.2 Practice recommendation #2 I found that tailored health care services and initiatives are being used throughout all participating health authorities, but it does not appear that these activities are being evaluated to show whether they made a difference in the health outcomes of the populations being served. At best, the services are established to demonstrate that the health authority is actively eliminating barriers to access for certain populations. I proposed that this situation is in part due to the limited number of strategies and evaluative frameworks that health authority administrators have in their toolkit when attempting to address access and health outcome differences and inequities. I believe that participants would benefit from having access to a larger toolkit of options for addressing health differences and inequities. I recommend that health authorities ought to invest in pilot projects specifically designed to test whether any new strategy explicitly designed to address population health differences or inequities demonstrate short- and mid-term outcomes. Long-term outcomes, however, would not be observable unless the pilot projects that were deemed successful were provided with ongoing funding, since long-term outcomes might not be seen in the population for ten or more years. This recommendation hinges on health authority administrators having access to a significantly increased amount of evaluation support services specifically focused on addressing how health services and initiatives are closing gaps in population health outcomes. 8.3.3 Practice recommendation #3 In relation to formal decision-making, I recommend that health authority  228 administrators ensure the following: (1) that any decision-making framework that is produced explicitly acknowledge decision tradeoffs and require decision-makers to examine their related effects; (2) that individual stakeholder value weighting processes of all selected decision-making criteria be integrated into the process, thus allowing participating stakeholders a better understanding of how shifts in personal priorities will influence the collective decision outcome; (3) that health authorities embed Daniels and Sabin’s accountability for reasonableness normative decision-making model by building into their decision-making processes formal ways for stakeholders to appeal decisions if so desired; and (4) that health authorities publicly publish and tell their communities which decision- making processes, principles and frameworks were used in decisions associated with selecting or eliminating health care services. 8.3.4 Practice recommendation #4 Currently, participating health authorities have shown varying levels of community involvement in their decision-making processes. Most often engagement was limited to the “inform” and “consult” end of the engagement continuum, however there were a few examples of sustained organized multiple level engagement strategies. Given participants’ stated support for public involvement in health care system decision-making, and participant and national government reports indicating the importance of moving away from an illness- focused health care system towards one that is wellness-focused (Romanow, 2002; Population Health and Wellness, 2005b), health authorities ought to do a better job at engaging all stakeholder groups throughout formal decision-making processes in order to  229 establish congruence between participant’s stated personal values and their decision-making process actions. Building in-house departments of public engagement at a regional planning and local operational level will increase the likelihood that health authority administrators will have the support they need to develop and implement evidence-based stakeholder engagement strategies. Depending on the issue, health authorities ought to consider increasing the amount of stakeholder involvement aligned with the “engage” end of continuum. Attempts to shift the health care system’s focus from one that is illness-based to one that is wellness focused requires a more significant discussion with the public given that it will have an effect on the types and amounts of services that they will be able to access. This is an issue that goes well beyond health authority control. As such, I recommend that the appropriate level of government fund public dialogical processes at a national and provincial level focused specifically on public perspectives regarding the desire to shift public resources away from the acute care system towards primary prevention, intervention and health promotion activities. Further, the public ought to be asked what safety measures ought to be established in order for them to feel comfortable with this proposed shift. 8.4 Future research recommendations 8.4.1 Research recommendation #1 Appreciative inquiry processes involve people who work within organizations to study and build upon what works well in their work setting (Bushe & Coetzer, 1995). This  230 combined quality improvement and qualitative research approach allows host organizations and researchers with an applied setting to study a phenomenon, develop an intervention, and evaluate the results. In BC, health authorities have a significant amount of experience with using this style of research. This is an excellent tool for researchers to gain access into applied research settings and to assist health care administrators located within BC health authorities in developing a peer-reviewed literature-informed decision-making tool that could be tested, improved, and spread across health authorities. Therefore I recommend that local researchers partner with health authorities to design and implement applied research related to the development of decision-making tools. Based upon my current findings, this proposed research on developing decision-making tools ought to include the examination of how administrators can weight personal values, make explicit important decision tradeoffs, and allow the tool to be embedded in everyday practice. Furthermore, this research ought to examine how the proposed tools address participating health administrator perspectives and concerns related with the potential loss of decision-latitude. 8.4.2 Research recommendation #2 Understanding how to effectively evaluate decisions that are implemented allows the decision-maker to reflect on and learn from his/her experiences. This valuable information assists the individual to build his/her personal mental models that helps him or her strengthen his/her decision-making expertise. However, there are many times where administrators decide not to approve or implement a request. In these instances, it is difficult to understand the impact of the decision or to examine how this relates to the development of mental  231 models. Judgemental heuristics operate at the individual level and allow people to collapse a large number of cognitive steps into clusters in order to speed up decision-making processes. Heuristics also assist individuals in assessing environmental cues for information that may help them come to a quick decision. In these cases, the judgemental heuristic can lead the individual to valid conclusions or may bias his/her direction. I recommend that further research focus on how to effectively evaluate the long-term effects of decisions that health care administrators do not approve. Furthermore, this research should explore how these types of decisions shape participants’ mental models and their development of a capacity for expert decision-making. Additionally, I recommend that research be performed in health care settings with administrators to more closely examine how heuristics frame and shape decision-making processes. The research design ought to draw significantly from the area of cognitive psychology and organizational theory. As identified in chapter seven, this research would benefit from a longitudinal research design that employs the use of participant observation and a series of follow-up interviews to discuss with the participant situations where they may have used a judgmental heuristic during the planning and decision-making process. I recommend that participant selection focus on those individuals who are known to their colleagues to be excellent decision-makers. Focusing on how these individuals come to quick decisions may provide others with a strong understanding of better ways of thinking through decisions and how they succeed in avoiding pitfalls associated with the inappropriate use of some heuristics.  232 8.5 Conclusion Participating health care administrators from four health authorities in BC provided their insights on what makes a good health care system, the usefulness of tailored health care services, and initiatives as a strategy for positively influencing sub-population health outcomes, as well as how they make decisions and how they know the decisions that they have made on the public’s behalf are good ones. I began this concluding chapter with a snapshot of participant perspectives on what it is like to work within a health region. This description was followed by a summary of my research findings that related to the four objectives that I documented in chapter one. Lastly, I ended this chapter with a series of practice and future research recommendations. Participants have offered a textured picture of the world of health care administration in BC health authorities. They regularly make difficult funding decisions that can have long-lasting effects on what health care services we can expect to receive. This is no easy space to inhabit. While many of their decisions may be unpopular, the courage and wisdom that they bring to their decision-making processes should be celebrated.  233 References Abelson, J., & Gauvin, F.-P. (2004a). Transparency, trust and citizen engagement: What Canadians are saying about accountability. Ottawa: Canadian Policy Research Networks Inc. Abelson, J., & Gauvin, F.-P. (2006). Assessing the impact of public participation: Concepts, evidence and policy implications. Ottawa: Canadian Policy Research Networks Inc. Abelson, J., Eyles, J., McLeod, C. B., Collins, P., McMullan, C., & Forest, P.-G. (2003a). Does Deliberation make a difference? Results from a citizens panel study of health goals priority setting. Health Policy, 66, 95-106. Abelson, J., Forest, P.-G., Eyles, J., Casebeer, A., & Mackean, G. (2004). Will it make a difference if I show up and share? A citizens' perspective on improving public involvement processes for health system decision-making. Journal of Health Services Research & Policy, 9(4), 205-212. Abelson, J., Forest, P.-G., Eyles, J., Smith, P., Martin, E., & Gauvin, F.-P. (2003). Deliberation about deliberative methods: Issues in the design and evaluation of public participation processes. Social Science & Medicine, 57, 239-251. Alvesson, M., & Sköldberg, K. (2000). Reflexive methodology: New vistas for qualitative research. Thousand Oaks: Sage. Arnstein, S. R. (1969). A ladder of citizen participation. Journal American Institute of Planners, 35, 215-224. Bazerman, M. H. (2002). Judgment in managerial decision making. (5th ed.). New York: John Wiley & Sons. Beauchamp, T. L., & Childress, J. (2001). Principles of Biomedical ethics. (5th ed.). New York: Oxford University Press. Beierle, T. C. (1998). Public Participation in Environmental Decisions: An evaluation framework using social goals. Washington: Resources for the Future. Böhm, G., & Brun, W. (2007). Intuition and affect in risk perception and decision making. Judgment and Decision Making, 3(1), 1-4.  234 Bottorff, J. L., Johnson, J. L., Irwin, L. G., & Ratner, P. A. (2000). Narratives of smoking relapse: The stories of postpartum women. Research in Nursing & Health, 23, 126-134. Bradford, N. (2005). Place-based public policy: Towards a new urban and community agenda for Canada. Ottawa: Canadian Policy Research Networks Inc. Braveman, P. (2006). Health disparities and health equity. Annual Review of Public Health, 27, 167- 194. Briggs, C. L. (2003). Interviewing, power/knowledge, and social inequality. In J. F. Gubrium & J. A. Holstein (eds.), Postmodern interviewing (pp. 243-254). Thousand Oaks, CA: Sage. British Columbia Medical Association. (2006). Doctors ratify six-year agreement. Vancouver: British Columbia Medical Association. British Columbia Ministry of Health. (2007a). Backgrounder: Health Innovation Fund breakdown. Victoria, BC: British Columbia Ministry of Health. British Columbia Ministry of Health. (2007b). Primary health care charter: A collaborative approach. Victoria, BC: British Columbia Ministry of Health. British Columbia Ministry of Health. (2008). Forum celebrates innovation in health care. Victoria, BC: British Columbia Ministry of Health. Buchanan, A. (1996). Toward a theory of the ethics of bureaucratic organizations, Business Ethics Quarterly, 6(4), 419-440. Bushe, G. R., & Coetzer, G. (1995). Appreciative inquiry as a team-development intervention: A controlled experiment. Journal of Applied Behavioral Science, 31(1), 13-30. Chafe, R., Neville, D., Rathwell, T., Deber, R., Kenny, N., Nestman, L., Joffres, C., Chappell, H., Conrad, P., & Batema, M. (2007). A framework for involving the public in healthcare coverage and resource allocation decisions. Halifax: Dalhousie University. Charmaz, K. (2002). Qualitative interviewing and grounded theory analysis. In J. F. Gubrium & J. A. Holstein (eds.), Handbook of interview research (pp. 675-694). Thousand Oaks: Sage. Chenail, R. J. (1997). Keeping things plumb in qualitative research. The Qualitative Report, 3(3), 1-7. Clandinin, D. J., & Connelly, F. M. (2000). Narrative inquiry: Experience and story in qualitative  235 research. San Francisco: Jossey-Bass. Clemen, R. T. & Reilly, T. (2001). Making hard decisions with decision tools. (2nd edn.) California: Duxbury. Coffey, A., & Atkinson, P. (1996). Making sense of qualitative data: Complementary research strategies. Thousand Oaks: Sage. Commission on Social Determinants of Health. (2007). A conceptual framework for action on the social determinants of health. Geneva: World Health Organization. Consumer Focus Collaboration. (2000). Improving health services through consumer participation. Canberra: Consumer Focus Collaboration Publication. Cox, S. M. (2003). Stories in Decisions: How at-risk individuals decide to request predictive testing for Huntington Disease. Qualitative Sociology, 26(2), 257-280. Creswell, J. W. (1998). Qualitative inquiry and research design: Choosing among five traditions. Thousand Oaks: Sage. Czarniawska, B. (2004). Narratives in Social Science Research. Thousand Oaks: Sage. Daniels, N., & Sabin, J. E. (2002). Setting limits fairly: Can we learn to share medical resources? New York: Oxford University Press. Daniels, N. (1985). Just health care. New York: Cambridge University Press. Daniels, N. (2001). Justice, health, and healthcare. American Journal of Bioethics, 1(2), 2-16. Daly, K. (1992). The fit between qualitative research and characteristics of families. In J. F. Gilgun, K. Daly, & G. Hadel (Eds.), Qualitative methods in family research (pp. 3-11). Newbury Park: Sage. Denzin, N.K. & Lincoln, Y. S. (Eds.) (2005). The discipline and practice of qualitative research (3rd ed.). Newbury Park: Sage. Doty, D. H., & Glick, W. H. (1994). Typologies as a unique form of theory building: Toward improved understanding and modeling. Academy of Management Review, 19(2), 230-251. Dryzek, J. S. (2000). Deliberative democracy and beyond: Liberals, critics, contestations. New York: Oxford University Press.  236 Fafard, P. (2008). Evidence and healthy public policy: Insights from health and political sciences. Quebec: National Collaborating Centre for Healthy Public Policy. Finlay, L. (2002). “Outing” the researcher: The provenance, process and practice of reflexivity. Qualitative Health Research, 12(4), 531-545. Fiorino, D. J. (1990). Citizen Participation and Environmental Risk: A survey of institutional mechanisms. Science, Technology, & Human Values, 15(2), 226-243. First Nations Leadership Council, G. o. C. a. G. o. B. C. (2007). Tripartite First Nations health plan. Flyvbjerg, B. (2001). Making social science matter: Why social inquiry fails and how it can succeed again. Cambridge: Cambridge University Press. Fontana, A. (2003). Postmodern trends in interviewing, In J. F. Gubrium & J. A. Holstein (eds.), Postmodern interviewing (pp. 51-65). Thousand Oaks: Sage. Frankish, C. J., Kwan, B., Ratner, P. A., Wharf Higgins, J., & Larsen, C. (2002). Challenges of citizen participation in regional health authorities. Social Science & Medicine, 54, 1471-1480. Frankish, C. J., Moulton, G. E., Quantz, D., Carson, A. J., Casebeer, A. L., Eyles, J. D., Labonte, R., & Evoy, B. (2007). Addressing the non-medical determinants of health: A survey of Canada's health regions. Canadian Journal of Public Health, 98(1), 41-47. Freeman, M. (1997). Why narrative? Hermeneutics, historical understanding, and significance of stories. Journal of Narrative and Life History, 7(1-4), 169-176. Fung, A. (2003). Association and Democracy: Between theories, hopes and realities. Annual Review of Sociology, 29, 515-539. Ganster, D. C., Fox, M. L., & Dwyer, D. J. (2001). Explaining employees' health care costs: A prospective examinations of stressful job demands, personal control, and physiological reactivity. Journal of Applied Psychology, 86, 954-964. Geanellos, R. (2000). Exploring Ricouer's hermeneutic theory of interpretation as a method of analysing research texts. Nursing Inquiry, 7, 112-119. Gibson, J., Martin, D., & Singer, P. (2005). Priority setting in hospitals: Fairness, inclusiveness, and institutional power differences. Social Science & Medicine, 16, 2355-2361.  237 Gladwell, M. (2006). Blink: The power of thinking without thinking. Penguin, USA. Government of British Columbia. (2005). Ministry of Health Services service plan 2005/06 - 2007/08. Victoria: Ministry of Health Services. Government of British Columbia. (1996). Health Authorities Act. Gubrium, J. F., & Holstein, J. A. (2003). Postmodern Interviewing. Thousand Oaks: Sage. Gutmann, A., & Thompson, D. (2004). Why Deliberative Democracy? New Jersey: Princeton University Press. Halma, L., Mitton, C., Donaldson, C., & West, B. (2004). Case study on priority setting in rural Southern Alberta: Keeping the house from blowing in. Canadian Journal of Rural Medicine, 9(1), 26-36. Hariri, N. (2003). Framework for community engagement. Vancouver: Vancouver Coastal Health. Health Canada. (1986). Achieving health for all: A framework for health promotion. Ottawa: Health Canada. Health Canada. (1999). Taking action on population health: A position paper for health promotion and programs branch staff. Ottawa: Health Canada. Herzog, H. (2005). On home turf: Interview location and its social meaning. Qualitative Sociology, 28(1), 25-47. Jacobs, L., Marmor, T., & Oberlander, J. (1999). The Oregon health plan and the political paradox of rationing: What advocates and critics have claimed and what Oregon did. Journal of Health Politics, Policy and Law, 24(1), 161-180. Jasanoff, S. (2005). Designs on nature: Science and democracy in Europe and the United States. New Jersey: Princeton University Press. Jewkes, R., & Murcott, A. (1998). Community representatives: Representing the "community"? Social Science and Medicine, 46(7), 843-858. Kahneman, D., & Tversky, A. (1996). On the reality of cognitive illusion. Psychological Review, 103(3), 582-591. Kaler, J. (2002). Responsibility, accountability and governance. Business Ethics: A European Review,  238 11(4), 327-334. Kaplan, M. S. (2004). Myths and realities of Canadian medicare. Oregon's Future, 5(2), 23-26. Kearney, R. (1994). Modern movements in European philosophy (2nd Ed.). Dover, NH: Manchester University Press. Keeney, R. L., & McDaniels, T. (1999). Identifying and structuring values to guide integrated resource planning at BC Gas. Operations research, 3(3), 396-405. Keeney, R. L. (1988). Structuring objectives for problems of public interest. Operations Research, 3(3), 396-405. Keeney, R. L. (1992). Value-focused thinking: A path to creative decision making. Harvard University Press. Klein, G. (1998). Sources of power: How people make decisions. Cambridge: The MIT Press. Klein, G. (2003). The power of intuition: Why developing your gut instincts will make you better at what you do. Doubleday, USA. Kosteniuk, J. G., & Dickinson, H. D. (2003). Tracing the social gradient in the health of Canadians: Primary and secondary determinants. Social Science & Medicine, 57, 263-276. Lavis, J. N. (2002). Ideas at the margin or marginalized ideas? Nonmedical determinants of health in Canada. Health Affairs, 21(2), 107-112. Lavis, J. N. (2006). Research, public policymaking, and knowledge-translation processes: Canadian efforts to build bridges, Journal of Continuing Education in Health Professions, 26(1), 37-45. Lincoln, Y. S., & Guba, E. G. (2000). Paradigmatic controversies, contradictions, and emerging confluences. In N. K. Denzin & Y. S. Lincoln (eds.), Handbook of qualitative research, 2nd edn. (pp. 163-188). Newbury Park: Sage. Litva, A., Coast, J., Donovan, J., Eyles, J., Shepherd, M., Tacchi, J., Abelson, J., & Morgan, K. (2002). "The public is too subjective": Public involvement at different levels of health-care decision making. Social Science & Medicine, 54, 1825-1837. Lomas, J. (1997). Devolving authority for health care in Canada's provinces: 4. Emerging issues and prospects. Canadian Medical Association Journal, 156(6), 817-823.  239 Lomas, J., Woods, J., & Veenstra, G. (1997). Devolving authority for health care in Canada's provinces: 1. An introduction to the issues. Canadian Medical Association Journal, 156(3), 371-377. Lukes, S. (1974). Power: A radical view. London: Macmillan. Macinko, J., & Starfield, B. (2002). Annotated Bibliography on Equity in Health, 1980-2001. International Journal for Equity in Health, 1(1), 1-20. Mansbridge, J. (2000). What does a representative do? Descriptive representation in communicative settings of distrust, uncrystallized interests, and historically denigrated status. In W. Kymlicka & W. Norman (eds.), Citizenship in diverse societies (pp. 99-123) New York: Oxford University Press. Martin, D., Abelson, J., & Singer, P. (2002). Participation in health care priority-setting through the eyes of the participants. Journal of Health Services Research & Policy, 7(4), 222-229. McDaniels, T. (2000). Creating and using objectives for ecological risk assessment and management, Environmental Science & Policy, 3, 299-304. McDaniels, T. (2000a). Creating and using objectives for ecological risk assessment and management, Environmental Science & Policy, 3, 299-304. McDaniels, T. (2000b). Using judgment in resource management: A multiple objective analysis of a fisheries management decision Operations research, 43(3), 415-422. McDaniels, T. L., Gregory, R. S., & Fields, D. (1999). Democratizing Risk Management: Successful Public Involvement in Local Water Management Decisions. Risk, 19(3), 497-510. McDonald, M. (2000). The governance of health research involving human subjects (HRIHS). Ottawa: Law Commission of Canada. McDonald, M. (2001). Canadian governance of health research involving human subjects: Is anybody minding the store? Health Law Journal, 9, pp. 1-21. Mendelsohn, M. (2002). Canada's social contract: Evidence from public opinion. Ottawa: Canadian Policy Research Networks. Menon, D., Starfinksi, T., Martin, D., Windwick, B., Singer, P., & Caulfield, T. (2003). Incorporating public values and technical information into health care resource allocation decision-  240 making. Edmonton: Alberta Heritage Foundation for Medical Research. Mitton, C., & Donaldson, C. (2002). Setting priorities in Canadian regional health authorities: A survey of key decision makers. Health Policy, 60, 39-58. Mitton, C., & Donaldson, C. (2003). Resource allocation in health care: Health economics and beyond. Health Care Analysis, 11(3), 245-257. Mitton, C., & Donaldson, C. (2004). Priority setting toolkit: A guide to the use of economics in healthcare decision making. London: BMJ Publishing Group. Mitton, C., Patten, S., Waldener, H., & Donaldson, C. (2003). Priority setting in health authorities: A novel approach to a historical activity. Social Science & Medicine, 57, 1653-1663. Mitton, C., Teng, F., Cranston, L., & MacKenzie, J. (2006). Priority Setting in the Provincial Health Services Authority: Case Study for the 2005/06 Planning Cycle. Healthcare Policy / Politiques de Santé, 2(1), 91-106. Morse, J. M., & Field, P. A. (1995). Qualitative research methods for health professionals, 2nd edn. Newbury Park, CA: Sage. Parliament of Canada. (2004). Canada Health Act. Payne, J. W., & Bettman, J. R. (1992). Behavioral decision research: A constructive processing perspective. Annual Review of Psychology, 43, 87-131. Pedersen, S., Barr, V., Wortman, J., & Rootman, I. (2007). 'Equity lens' evidence review: Effective interventions to reduce health inequities. Public Health Association of BC, British Columbia. Pfister, H.-R., & Böhm, G. (2008). The multiplicity of emotions: A framework of emotional functions in decision making. Judgement and Decision Making, 3(1), 5-17. Population Health and Wellness, M. o. H. S. (2005a). A framework for core functions in public health. Victoria: Province of British Columbia. Population Health and Wellness, M. o. H. S. (2005b). Public health renewal in British Columbia: An overview of core functions in public health. Victoria: Province of British Columbia. Rawls, J.  (2003).  A theory of Justice (Revised ed.). United States: Oxford University Press. Riessman, C. K. (1993). Narrative analysis. Newbury Park, CA: Sage.  241 Roediger, J. L. (1990). Implicit memory: Retention without remembering. American Psychologist, 45(9), 1043-1056. Romanow, R. J. (2002). Building on Values: The future of healthcare in Canada — Final report. Rowe, G., & Frewer, L. J. (2005). A typology of public engagement mechanisms. Science, Technology, & Human Values, 30(2), 251-290. Rowe, G., Marsh, R., & Frewer, L. J. (2004). Evaluation of a deliberative conference. Science, Technology & Human Values, 29(1), 88-121. Ruger, J. P. (2007). Rethinking Equal Access: Agency, Quality, and Norms. Global Public Health, 2(1), 78-96. Sandelowski, M. (1993). Rigor or Rigor Mortis: The Problem of Rigor in Qualitative Research Revisited. Advances in Nursing Science, 16(2), 1-8. Sandelowski, M. (1995). Focus on qualitative methods. Triangles and crystals: On the geometry of qualitative research. Research in Nursing & Health, 18, 569-574. Sandelowski, M. (2000). Whatever happened to qualitative description? Research in Nursing & Health, 23, 334-340. Smythe, W. E., & Murray, M. J. (2000). Owning the story: Ethical considerations in narrative research. Ethics & Behavior, 10(4), 311-336. Strauss, A., & Corbin, J. (1998). Basics of qualitative research: Techniques and procedures for developing grounded theory. Thousand Oaks, CA: Sage. Terry, L. M. (2004). An integrated approach to resource allocation. Health Care Analysis, 12(2), 171- 180. Todd, P. M., & Gigerenzer, G. (2000). Précis of simple heuristics that make us smart. Behavioral and Brain Sciences, 23, 727-741. Vancouver Coastal Health Richmond Health Services Ethics Committee (2002). Toward a just distribution of scarce resources and the burdens of cut-backs in health care. von Winterfeldt, D. (1992). Expert knowledge and public values in risk management: The role of decision analysis. In S. Krimsky & D. Golding (eds.), Social Theories of Risk (pp. 321-342).  242 Westport, CN: Praeger. Wharf Higgins, J. (1999). Closer to Home: The case for experiential participation in health reform, Canadian Journal of Public Health, 90(1), 30-34. Whittemore, R., Chase, S. K., & Mandle, C. L. (2001). Validity in qualitative research. Qualitative Health Research, 11(4), 522-537. Wieland, J. (2001). The ethics of governance. Business Ethics Quarterly, 11(1), 73-87. Wynne, B. (2001). Creating public alienation: Expert Cultures of risk and ethics on GMOs. Science as Culture, 10(4), 445-481.   243 Appendix A: Individual Interview Protocol An examination of the ethical decision-making processes used in decisions to fund or cease funding tailored health services Topics for individual interviews with health region administrators who have had lead responsibility in deciding to fund (reduce or eliminate funding to) a community-based health service designed to meet the unique health needs of a subgroup population. Participating health region administrators will be asked to speak about their experiences with having to decide whether to fund, reduce, or cease funding a community- based service designed to meet the health-related needs of a subgroup population. They will be encouraged to share their perspective on procedural issues, including the structures used to involve other people in the decision making process (e.g. committees), and various inputs into the decision-making proces (e.g. financial, literature review, statistics). New issues may be raised during the piloting of this interview protocol [n=3] that may flag the necessity to ask new questions with the remaining interview participants. These interviews will be semi- structured in nature in that questions will be asked but the interview will be flexible enough to allow the participant to raise what s/he feels are important issues in relation to the topic. 1. Please tell me about your role in the health region. Sample Prompts: How long have you been working in health administration? How often are you involved in resource allocation processes? 2. Tell me about your thoughts on what makes a good health care system. Sample Prompts: What kind of values would drive the system? How would it function? What would it include?  244 3. Often services are created to directly address a specific sub-population like women’s clinics, Aboriginal outreach services, and youth addictions counselling as a way to address health disparities. What are your general thoughts about this strategy of making tailored services for specific groups? Sample Prompts: How do you feel about such programs versus mainstream programs? Do you see them making a difference in relation to providing better care to these groups? Can you describe any general patterns in the community and health region that precede a decision to fund a new service set up to serve a sub-population? 4. Please describe a situation where you were responsible for making a decision about funding (or reducing or cutting funding to) a community-based health service meant to serve a subgroup population. Sample Prompts: How did this situation come to your attention? Who was involved in the situation? Why did you or your organization identify this as a situation worth addressing? 5. Please describe how you went about the decision-making process. Sample Prompts: What kind of structures did you use (e.g. committee structure, meetings)? Who was involved in the process? Were there people from outside of the health authority involved (e.g. community members, non-profit agency staff)? Were there minutes used to track process? 6. What kinds of things did you feel were important elements in making your decision? Sample Prompts: How were different sources of data used such as financial, research literature, health-  245 related statistics, stakeholder testimony, ethical or moral considerations, etc? 7. Tell me more about X (identified input). How did you use this information? Walk me through the process of weighing its importance on the decision to be made. Sample Prompts: How did you come to decide that this information was of a good enough quality to include into your decision-making process? Were there any pieces of information that you felt would have been useful that was not available to you? If so, what were they? 8. Based on the decision-making inputs that you described, what were the most important pieces of information to have in order to make an effective decision? Sample Prompts: What makes these important to you? What would happen if one or two of these were not available to you? 9. Are there things that tell you that you’ve made a good decision? If so, what are they? Sample Prompts: Substantive considerations such as adhering to principles of justice? Procedural consideration such as the way a decision was made? How other people responded to the decisions?  246 Appendix B: Individual Consent Form An examination of the ethical decision-making processes used in decisions to fund, reduce or cease funding tailored health services Brian Evoy (Graduate Student) (604.708.4314) University of British Columbia Dr. Jim Frankish (Graduate Student Supervisor) (604.822.9205) Consent Form Individual Interviews with Health Region Administrators Purpose of Study: The purpose of this research is to examine how healthcare administrators decide to fund, reduce, or cease funding community-based health care services developed for parts of the population that require tailored services to meet their unique needs. Examples of these services include women’s clinics, Aboriginal healing centres, and youth addictions services. This study will gather information from documents, interviews with 24 health region administrators. Research Objectives include the following: 1. Understand how participant perceptions are experienced and organized. 2. Identify the criteria that health region administrators and non-government agency administrators perceive to be important in health authority resource allocation decision-making processes. 3. Describe the ethical dimensions that interview participants attribute to the decision- making process. 4. Examine how participant descriptions of how decisions are made align with existing decision-making frameworks (decision analysis, naturalistic decision models, and ethical decision-making models) and add to this area of knowledge.  247 Study Procedures: The study involves two methods of information gathering: 1. Individual interviews with health administrators 2. Relevant documents (e.g. meeting minutes, reports) As a health region administrator who has been in a position to fund, reduce, or cease funding a community-based tailored health care service, you are being invited to participate in one 1.5-hour individual interview. You will be asked about your experiences with, and understanding of, allocating health care resources in relation to a specific community-based health care service that you choose to discuss. You will be asked about your thoughts related to the strategy of designing specific services to sub-group populations as a way address health disparities, to describe the process and structures you used in your decision-making process, and to identify the most important inputs necessary to make an effective resource allocation decision. Brian Evoy (Graduate Student) will facilitate the 1.5-hour interview in a location and time that is convenient to you. A follow-up call may be necessary to clarify issues arising from this meeting. Your Participation: Your participation in this study is entirely voluntary and you may refuse to participate or decide to withdraw at any time. After the interview, you will be invited to review any quotes that you have provided and the researcher intends to use in a public document. You will be asked to assess whether your privacy has been sufficiently protected, and whether the way the quote is being used aligns with the original context within which you provided the quote. The time you commit to this research will range from 1.5 hours to 3 hours (includes any follow up questions or reviewing quotes). Beyond specific quotes that are used to provide evidence for the researcher’s interpretations, findings will only be reported in a combined form. Individual interview findings will not be shared with others outside of the Graduate Student supervisory committee. Confidentiality: All information will be kept strictly confidential. Any kind of identifying information, including names, cities/towns, organizations or personal descriptions, will be altered. All audiotapes, transcripts, consent forms, and documents will be identified by code and kept in a locked file cabinet in the graduate student’s researcher’s office. The electronic version of these documents will be locked down with a password. The only people who will hear the  248 unaltered audiotapes or transcripts will be the graduate student (Brian Evoy), a transcriptionist, and his supervisory committee. All audiotapes will be destroyed five years after completion of the Ph.D. thesis. Research findings from meeting with health care administrators in four geographically-based BC health regions will be combined with document obtained through the health region administrators, analyzed and written up to form the basis of the graduate student’s Ph.D. thesis and publications. These will be shared through presentations with health region personnel. You may request copies of the thesis or publications about the findings of the study. Upon completion of this study, the graduate student may conduct further analysis this data set related to health region uses of public involvement strategies in decision-making processes. The anonymized transcripts, field notes an other research documents will not be destroyed. With your consent, they will be kept by the the graduate student (Brian Evoy) in electronic form under a protection password. All future uses of this data must be approved by a Research Ethics Board. Risks and Potential Benefits: No anticipated risks are related to your involvement in this study. Given the nature the interview, topics may be raised that you may find uncomfortable to discuss. It is entirely up to you to answer questions or express your views. The facilitator will respect your decisions. There may be benefits to your involvement in relation to assisting in the development of knowledge that may support health region administrators like yourself to make resource allocation decisions transparent for review by members of the public. Remuneration/Compensation: Your participation in this study is entirely voluntary and you will not be compensated financially. However if you incur personal costs associated with participating, you will be reimbursed. Costs include travel, parking, and childcare. Contact: Your participation in this study must be entirely voluntary and based on a clear understanding of the research. Please contact Brian Evoy (Graduate Student) at 604.708.4314 or Dr. Jim Frankish (Graduate Student Supervisor) at 604.822.9205 to seek additional information. If you have any concerns about your rights or treatment as a research subject,  249 you may contact the Director of Research Services at 604.822.8598. Consent: My signature below indicates that: 1. I have received an explanation of the study and that all of my questions have been answered to my satisfaction. 2. I understand that my participation is entirely voluntary and that I may withdraw from the study at any point in the research project. 3. I agree to participate in a 1.5 hour interview. 4. I have been given a signed copy of the consent form for my personal records. In addition, by checking the appropriate box below, F I agree that the research data and documents that have had all participant-identifying information removed may be kept by the Graduate Student in a secure location for an indefinite period of time for the purpose of conducting future post doctoral research related to health region uses of public involvement strategies in decision-making processes. F I do not agree to that the research data and documents that have had all participant- identifying information removed may be kept by the Graduate Student in a secure location for an indefinite period of time for the purpose of conducting future post doctoral research related to health region uses of public involvement strategies in decision-making processes and therfore request that all such documents be destroyed 5 years after the completion of the Ph.D. thesis. Signature: ________________________________ Date Signed: __________________  250 Appendix C: Chapter Structure of Results Chapters Chapter four  251 Chapter five   252 Chapter six  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items