Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Progressively engaging : how nurses, patients and family members manage relationships in acute care hospital… Segaric, Cheryl Ann 2007

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


831-ubc_2007-319246.pdf [ 23.32MB ]
JSON: 831-1.0100783.json
JSON-LD: 831-1.0100783-ld.json
RDF/XML (Pretty): 831-1.0100783-rdf.xml
RDF/JSON: 831-1.0100783-rdf.json
Turtle: 831-1.0100783-turtle.txt
N-Triples: 831-1.0100783-rdf-ntriples.txt
Original Record: 831-1.0100783-source.json
Full Text

Full Text

PROGRESSIVELY ENGAGING: HOW NURSES, PATIENTS AND FAMILY MEMBERS MANAGE RELATIONSHIPS IN ACUTE CARE HOSPITAL SETTINGS by CHERYL ANN SEGARIC B.ScN., The University of Windsor, 1983 M.ed., Simon Fraser University, 1993 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Nursing) THE UNIVERSITY OF BRITISH COLUMBIA August 2007 © Cheryl Ann Segaric, 2007 ii ABSTRACT This grounded theory study, informed by symbolic interactionism, explains how nurses, patients, and family members manage relationships in order to plan and provide care in acute care hospital settings. The study also explains the effects of contextual and systemic features associated with acute care environments and participants' personal characteristics on their efforts to manage relationships. Data collection included thirty three hours of participant observation and forty interviews. Seventeen interviews were conducted with nurses, ten with family members, and thirteen with patients. Participants were recruited from a total of ten acute care units across four community hospitals in the Fraser Health Authority of British Columbia; there were four medical units, three surgical units, two transitory care or activation units, and one community hospital intensive care unit. I constructed the basic social psychological process of progressively engaging. The process describes how nurses, patients, and family members manage their relationships during patient care by developing varying levels of engagement. The levels of engagement are represented by three stages, including: focusing on tasks, getting acquainted, and building rapport. Structural conditions and personal factors, relevant to nurses, patients, and family members, facilitate or constrain their efforts to progressively engage by contributing to or detracting from their shared perspectives. Levels of engagement ranged from 'just doing the job' to 'doing the job with heart' or making a deep human connection. Higher levels of engagement achieved in nurse, patient, and family member relationships correspond with more satisfaction expressed by participants about their relationships and nursing care. The substantive theory of progressively engaging makes a significant contribution to the family nursing theory. The process has implications for nursing education, practice, research, and administration. iii TABLE OF CONTENTS Abstract ii Table of Contents iii List of Tables viii List of Figures ix Acknowledgements x Dedication xi 1 Introduction. 1 1.1 Background 1 1.2 Significance 3 1.3 Statement of Problem 5 1.4 Statement of Purpose 6 1.5 Research Questions 7 1.6 Conceptual Definitions 7 1.6.1 Family 7 1.6.2 Acute care 8 1.7 Overview of Chapters 8 2 Literature Review 9 2.1 Issues Related to Language 10 2.1.1 Lack of a common language 10 2.1.2 Theory to practice gap 11 2.1.3 Family as context 12 2.1.4 Family as unit 13 2.1.5 Family system 14 2.1.6 The need for conceptual clarity 16 2.2 Issues Related to Contextual Features 18 2.2.1 Failure to account for the work environment 19 2.2.2 Lack of nursing time 21 2.2.3 Implications of time for nurse-family relationships 23 2.2 4 Other contextual issues affecting nurse-family interaction 25 2.3 Issues Related to Interactive Processes 28 2.4 Summary 36 iv TABLE OF CONTENTS continued. 3 Theoretical Framework: Symbolic Interactionism 37 3.1 An Overview of Symbolic Interactionism 38 3.2 Pragmatisim ....39 3.3 Ontology 41 3.4 Epistemology 45 3.5 Methodology 50 3.6 Summary 54 4 Strategy of Inquiry: Grounded Theory 56 4.1 Grounded Theory as Informed by Symbolic Interactionism 56 4.2 Grounded Theory - An Overview 58 4.3 Research Design 61 4.3.1 Purpose 61 4.3.2 Questions 62 4.3.3 Initial purposive sampling 63 4.3.4 Inclusion/exclusion criteria for patient participants 65 4.3.5 Inclusion/exclusion criteria for nurse participants 66 4.3.6 Inclusion/exclusion criteria for family member participants 67 4.4 Ethical Considerations 67 4.4.1 Obtaining ethical approval 69 4.4.2 Ethical approval extension 70 4.5 Procedures 70 4.5.1 Negotiating access 70 4.5.2 Participant recruitment strategies 71 4.5.3 Data collection 73 4.5.4 Descriptive data 74 4.5.5 Semi-structured interviews 75 4.5.6 Audio-taping 77 4.5.7 Participant observation 79 4.5.8 Field notes 82 4.5.9 Personal journal 84 4.6 Data Analysis 85 4.6.1 Open coding 86 4.6.2 Selective coding for a core category 91 4.6.3 Core category 93 4.6.4 Theoretical coding and sorting 94 4.6.3 Saturation and completeness 95 4.7 Criteria for Rigor 97 4.8 Limitations 103 4.9 Summary 104 V TABLE OF CONTENTS continued. 5 Study Findings 106 5.1 Sample 106 5.1.1 Description of patient participants 106 5.1.2 Description of nurse participants 108 5.1.3 Description of family member participants 110 5.2 Progressively Engaging: Theoretical Overview 112 5.3 Stages of Engaging 114 5.3.1 Stage one: Focusing on tasks 116 5.3.2 Stage two: Getting acquainted 119 5.3.3 Stage three: Establishing rapport 120 5.4 Grounded Theory: The Progressively Engaging Process 121 5.4.1 Stage one: Focusing on tasks 121 5.4.2 Structural conditions and personal factors 122 5.4.3 Terms of engagement 124 5.4.4 Perception of time 126 5.4.5 Language barriers 128 5.4.6 Patient/family cooperation/compliance 130 5.5 Interpersonal Dynamics 131 5.5.1 Positive interpersonal dynamics 132 5.5.2 Negative interpersonal dynamics 134 5.6 Falling Through the Cracks 138 5.6 Feeling Each Other Out 143 5.7 Stepping Back 145 5.8 Just Doing the Job as Outcome 150 5.9 Section Summary 153 5.10 Stage Two: Getting Acquainted 154 5.10.1 Terms of engagement: Business 156 5.10.2 Terms of engagement: Acuity 158 5.10.3 Terms of engagement: Continuity of care 161 5.10.4 Terms of engagement: Family involvement 163 5.10.5 Terms of engagement: Friendliness 165 5.10.6 Terms of engagement: Positive interpersonal dynamics 166 5.10.7 Terms of engagement: Questions 168 5.10.8 Terms of engagement: Personal sharing 171 5.10.9 Terms of engagement: Cooperative /compliant patients and families 174 5.10.10 Terms of engagement: Negative interpersonal dynamics and stepping back.. 174 5.11 Section Summary 177 5.12 Stage three: Establishing Rapport 178 5.12.1 Respect, trust, and reciprocity 179 5.12.2 Interpersonal dynamics 180 5.12.3 Stepping back 187 5.12.4 Doing the job with heart as outcome.. 188 5.13 Chapter Summary.. 191 TABLE OF CONTENTS continued. 6 Discussion and Implications of the Study Finding 192 6.1 Progressively Engaging: A Summary 192 6.2 Progressively Engaging as a Contribution to Nursing Theory 194 6.3 Methodological Contribution 196 6.4 Conceptualization of Family 198 6.4.1 Pragmatism 206 6.5 Accounting for Systemic and Contextual Features of the Work Environment 208 6.5.1 Time 210 6.5.2 Acuity 216 6.5.3 Increases in casual nursing staff. 217 6.5.4 Influences of structural conditions on personal factors 218 6.6 Interactive Processes 220 6.6.1 Fast-tracking 223 6.6.2 Stepping back 224 6.6.3 Using questions 226 6.6.4 Reciprocity 229 6.6.5 Shared responsibility 233 6.6.6 Social capital 235 6.7 Implications 236 6.7.1 Implications for nursing practice 236 6.7.2 Implications for nursing administration 243 6.7.3 Implications for nursing education 246 6.7.4 Implications for nursing research 250 6.8 Limitations 255 6.9 Conclusion 256 6.10 Chapter Summary 257 Bibliography 258 Appendices 269 Appendix A - The University of British Columbia Office of Research Services and Administration Behavioural Research Ethics Board Certificate of Approval 269 Appendix B - Fraser Health Authority Clinical Investigation Committee Research Approval Letter 271 Appendix C - South Fraser Health Authority Regional Research Review Committee Research Approval Letter 274 Appendix D - Fraser Health Authority Clinical Investigation Committee Approval for Extension 277 vii TABLE OF CONTENTS continued. Appendix E - Participant Information Letters 279 Appendix F - Consent Forms 288 Appendix G - Demographic Questionnaire for Nurse Participants 297 Appendix H - Demographic Questionnaire for Family Member Participants 300 Appendix I - Demographic Questionnaire for Patient Participants 303 Appendix J - Initial Interview Guide for Use with Nurse Participants 306 Appendix K- Initial Interview Guide for Use with Patient Participants 309 Appendix L - Initial Interview Guide for Use with Family Participants 312 Appendix M - Participant Observation Guide 315 Appendix N - Sample Field Notes and Participant Observation Notes 318 Appendix O - Sample Conceptual Diagrams 328 Appendix P - Sample Memos 334 viii LIST OF TABLES Table 1.1 Demographic Characteristics of Patient Participants 107 Table 1.2 Demographic Characteristics of Nurse Participants 109 Table 1.3 Demographic Characteristics of Family Member Participants I l l ix LIST OF FIGURES Figure 1.0 Progressively Engaging Process 115 Figure 2.0 Stages of Progressively Engaging 118 ACKNOWLEDGEMENTS I would like to thank the nurses, patients, and family members who participated in this study. Without them this study would not have been possible. I would also like to thank the nurse managers, clinical resource nurses, and staff nurses from the participating acute medical and surgical units for supporting my efforts to understand how nurses, patients, and family members manage their relationships in these settings and specifically for their assistance in the recruiting process. To my supervisory committee members Dr. Angela Henderson and Dr. Carol fillings: thank you for the extremely insightful comments and feedback you provided at various stages throughout the research process. It has been a pleasure and an honor to have had the opportunity to work with you and learn from you. I very much appreciate the time and effort that you have spent on my behalf as well as your ongoing support and encouragement. Finally, to my research supervisor Dr. Wendy Hall: words cannot express my gratitude for all that you have done in an effort to assist me throughout the research process. You inspired, encouraged, and supported me in my effort to reach for and ultimately achieve goals far beyond what I ever imagined I could. It has been such an honor and a privilege to work with you. Your dedication to the nursing profession - practice, education, and research is beyond compare and I am proud to call you my mentor. Thank you for everything! DEDICATION I dedicate this work to my family: Tony, Craig, and Stephanie Segaric. The Friend Who Stands By When trouble comes your soul to try, You love the friend who just stands by. Perhaps there's nothing he can do; The thing is strictly up to you. For there are troubles all your own, And paths the soul must tread alone; Times when love can't smooth the road, Nor friendship lift the heavy load. But just to feel you have a friend, Who will stand by until the end; Whose sympathy through all endures, Whose warm handclasp is always yours. It helps somehow to pull you through, Although there's noting he can do; And so with fervent heart we cry, "God bless the friend who just stands by" (author unknown) It was you, my loving husband and children, who stood by me every step of the way; from beginning to end. I could not have completed this journey without your unwavering support and understanding. I love you with all my heart. Thank you. 1 CHAPTER ONE 1 Introduction For the purpose of establishing the sociological perspective for this research study, this chapter begins with a brief overview of some of the key issues confronting the British Columbian health care system, particularly as they relate to the acute care hospital setting. A description of the research questions is also provided. Key terms relevant to the study are defined. Finally, a brief overview of the chapters comprising this dissertation is presented. 1.1 Background It has been well publicized in recent years that the health care system in British Columbia and nurses in particular are faced unprecedented challenges around providing quality health care to the people of British Columbia (B.C.). Indeed, the B.C. health care system continues to be threatened by such issues as escalating health care costs, changing social demographics, for example an aging population and subsequent changing trends in general health care needs, a declining nursing workforce, and deteriorating working conditions for nurses (Canadian Institute of Health Information, 2002; Select Standing Committee Report on Health, 2001). Since the early 1990s, these various challenges have served as catalysts for an assortment of restructuring and downsizing initiatives across the provincial health care system including acute care bed and hospital closures, an increase in outpatient services, and the reduction of hospital admissions and length of stay. While these actions have yet to be evaluated, the consequences are being felt by nurses within acute care hospital environments in the form of higher in-patient acuity, longer waiting times for diagnostic and treatment interventions, increased difficulty obtaining immediate care, and increased burden of care for family members both in and outside the 2 hospital (British Columbia Ministry of Health Planning, 2002; Canadian Institute for Health Information, 2002). The working conditions of nurses have also been affected - especially those nurses practicing in acute care hospitals. The most common concerns include: the burden of heavy patient loads (in terms of both numbers and acuity), the number of non-nursing tasks required of nurses, and the lack of administrative support and leadership. In a qualitative analysis of comments by 2,500 hospital nurses regarding career and workplace, Dunleavy, Shamian and Thomson (2003) noted that nurses are angry and frustrated at a system that seems to have forgotten that health care is about the people. Dunleavy, et al. writes: "...there is growing concern that nurses are reaching a point of overload, where they will no longer be able to put patients first" (p. 25). Ironically, putting patients and their families first is the mandate given nurses across B.C. by the College of Nurses of British Columbia (Professional Standards for Registered Nurses and Nurse Practitioners). While issues such as increased patient acuity, reduced length of hospital stay, increased numbers of family members in crisis, and increased nursing workloads accentuate the importance of nursing care that includes families in health care delivery, there is evidence that family-oriented nursing care has become more elusive in acute care practice settings. The disparity between the importance of family nursing and its apparent lack of implementation raises several questions around the relationships between the complexities of the contemporary acute care hospital environment and the establishment of collaborative relationships among nurses, patients, and family members. Indeed, an underlying assumption of this research study is that experiences and perceptions associated with various challenges in the system and subsequent 3 changes in health care policy and delivery of services influence interactive processes among nurses, patients and family members in terms of planning and providing care. 1.2 Significance While understanding the key challenges affecting the health care system generally and acute care hospitals specifically is important, the reciprocal effects of the health care system and the people it serves must also be considered. With escalating costs prompting the reorganization of health care delivery combined with increased demands for health care services, the role of families in health care has taken on new meaning and significance. Although the family unit has always been recognized as an important resource for promoting and maintaining the health of its members, today's families are also expected to assume increased responsibility for providing care to ill family members both in hospital and upon discharge (Conway-Giustra, Cowley & Gorin, 2002; Lee & Craft-Rosenberg, 2002; Wright & Leahey, 2000). This expectation comes at a time when contemporary families are experiencing increased complexity in their structure, development, roles, and functions (Friedman, Bowden & Jones, 2003). In the unexpected event of a family member experiencing an acute illness episode, disrupted routines, altered roles, and strain placed on financial and other family resources, secondary to the addition of the family care-giving role, may result (Denham, 2003; Friedman, Bowden & Jones, 2003). As noted by Duhamel (2004) in response to illness of a family member the entire family may be "shaken functionally and emotionally, and the stability of everyday life is threatened" (p. 68). This in turn, may lead to considerable distress within the family and ultimately undermine both the health of the patient and the family (Denham, 2003; Wright & Leahey, 2005). While there is clear evidence that families are assuming increased responsibility for pre and post-hospital care (Chisholm, 2000), it is less clear how family members are involved in care 4 provided to kin who are hospitalized for acute illness. Some study findings report extensive involvement by family in patient care, particularly with regard to emotional support and meeting basic physical needs (Astedt-Kurki, Paunonen & Lehti,1997); however, many studies have found that family participation is minimal or sporadic (Astedt-Kurki, Lehti, Paunonen, & Paavilainen, 1999; Gavaghan & Caroll, 2002; Paavilainen, Seppanen, & Astedt-Kurki, 2001). Several benefits are associated with participation of family in care, for example, better outcomes for patients promoting health of the family in general (Robinson & Wright, 1995; Wright & Leahey, 2000); enhanced feelings of empowerment among patients and families (Allen, 2000; Newton, 2000); diminished feelings of helplessness by patients and families; and improved compliance and earlier hospital discharge (Allen, 2000). The costs of family involvement in patient care can include guilt, fear, worry, tension, anger, frustration, feelings of powerlessness, role conflict, uncertainty about the future, economic burden, negative attitudes towards the condition or illness, lack of confidence in the family's ability to cope, and impaired health status of family members (Conway-Giustra, Crowley & Gorin, 2002; Fleury & Moore, 1999; Foxall & Gaston-Johansson, 1996; Lee & Craft-Rosenberg, 2002; Yates, 1998). Such effects are more pronounced for those family members who bear the primary burden of informal care giving. In the face of increased complexity of roles, structures, and functions in the family, secondary to such social trends such as delayed childbearing, aging parents, and dual-earner families, there has been little recognition of shifts in family demographics and roles by society in general and the health care system in particular (Jacobs & Gerson, 2001; Loveland-Cherry, 1996). Shorter hospitalizations for acute illness have contributed to shorter periods of contact for health care providers and patients and families, less time for discharge assessment, and limited negotiation with family members around discharge to home (Robinson and Wright, 1995). Family members, 5 particularly women and those with few resources and limited expertise, experience more anxiety about their responsibility to care for an acutely ill family member (Denham, 2003; Ward-Griffin & McKeever, 2000). During an acute illness episode, additional family care-giving responsibilities may result in disrupted routines, altered roles, and strain placed on financial and other family resources which may lead to considerable family distress and poorer patient and family member health outcomes (Denham, 2003). Consideration of the impact of illness on the whole family has directed attention to the need for nurses to expand their focus of care to be more inclusive of family. This has prompted a proliferation in family nursing publications, educational programs, and family nursing research aimed at advancing family theory for practice (Vosburgh & Simpson, 1993). In spite of these efforts, it has been widely observed that only limited progress in the transfer of family nursing theory to clinical nursing practice has occurred. As Duhamel (1995) stated "In too many clinical settings family nursing is almost nonexistent or developing very slowly" (p. 7). More recently, Hanson (2005) argued that "family nursing is fading despite the fact that families continue to be our best resource for health care" (p. 336). This theory-practice gap has been explained by the lack of specific descriptions of nursing practice with families (Bell, 1995), a lack of interface between nursing practice theory and family nursing theory (Vaughan-Cole, 1998), and various contextual factors that reduce nursing time (Wright & Leahey, 2005). 1.3 Statement of Problem Nurses in British Columbia are mandated to focus nursing care on patients and their families (Professional Standards for Registered Nurses and Nurse Practitioners). This directive is based primarily on extant knowledge and theory related to the reciprocal mfluence of families' health/illness status on all family members (Friedman, Bowden & Jones, 2003; Wright & 6 Leahey, 2005). Numerous barriers to implementation of family theory and to the formation of collaborative relationships among nurses, patients, and family members exist in acute care hospital settings. While barriers and challenges may be contextual and systemic in nature, little is understood about nurse-patient-family relationships as they develop and evolve in contemporary acute practice settings. Many nursing textbooks present family theory and nursing care approaches; however, few nurses have described or explained the interactions between nurses, patients, and families in acute care clinical practice (Vaughan-Cole, 1998). Consequently, the nature and form that family nursing in the acute care setting takes from the various perspectives of nursing, patient, and family is not known. 1.4 Statement of Purpose The purpose of my dissertation research entitled: Progressively Engaging: How nurses, patients and family members manage relationships in the acute care hospital setting, is to explain how nurses, patients, and family members who come together in acute care hospital settings manage relationships in order to plan and provide care. A secondary purpose is to explain the effects of contextual, systemic and other features associated with the acute care environment on efforts to manage relationships for the purpose of planning and implementing care. Finally, a third purpose is to explain how features associated with the acute illness and hospitalization experience that are of a more personal nature (e.g. beliefs and values), affect efforts to manage relationships among nurses, patients, and families. It is anticipated that the theory emerging from this study will provide explanations about the unique circumstances in acute care settings that influence nurses', patients', and family members' efforts to manage relationships during the provision of care for patients and their families. The findings are also anticipated to suggest nursing policy and education initiatives 7 aimed at promoting and supporting nurses' collaborative relationships with patients and families experiencing acute illness. 1.5 Research Questions In the context of implementation of family nursing in the acute care setting, the general area of concern associated with this dissertation research is expressed as understanding the process of managing relationships, between nurses, patients, and family member(s). The overall goal is to explain how nurses, patients, and family members manage relationships in order to plan and implement care. The specific research questions are stated as: 1. How do nurses, patients, and family members perceive their management of relationships in order to plan and provide patient care in the acute care hospital setting? 2. How do contextual and systemic features or structural conditions affect nurses', patients', and family members' efforts to manage relationships to plan and provide patient care? 3. How do personal factors affect nurses', patients', and family members' perceptions of their efforts to manage relationships to plan and provide patient care? 1.6 Conceptual Definitions 1.6.1 Family. For the purpose of this research study, family is broadly defined with a focus on relationships and interactions. Family, therefore, includes two or more persons interacting by virtue of kinship, friendship or shared living space, who may or may not be related by blood or marriage but who are joined by bonds of sharing and emotional closeness (Bell, 1995; Foxall & Gaston-Johansson, 1996; Friedman, Bowden & Jones, 2003). In terms of participation in the study, family members will include those persons identified by the patient as key participants or sources of support in the process of care planning and/or delivery. g 1.6.2 Acute care. For the purpose of this study, acute care is defined as a level of care involving short-term hospitalization for patients experiencing acute illness, injury, or exacerbation of a disease process (Flintoff & Williams, 1998). Patients experiencing acute illness or injury episodes are those that require specialized treatment, medication and coordinated professional assessment and care until such time that their condition is stabilized. 1.7 Overview of Chapters This chapter identified the sociological perspective or context of the study by providing a brief discussion regarding the background and significance of the phenomenon of interest. A description of the problem, the purpose, the research questions, and definitions of terms used in the study was also provided. Chapter two provides a synthesis of the literature addressing the difficulties and gaps associated with application of family nursing theory to practice. Chapter three describes the theoretical framework used to guide this interpretive research study. Chapter four offers a detailed account of the grounded theory method and procedures for data collection and analysis utilized in this study, as well as rigor and ethical considerations. In chapter five, the study findings are presented including a description of the sample and the theory of progressively engaging - the basic social psychological process which was constructed from the data that describes how nurses, patients, and/or family members manage relationships in the provision of care during acute care hospitalization. Finally, chapter six includes a discussion of the findings, as well as the implications of the study findings for nursing education, practice and family nursing theory development. Recommendations for further research are also considered. 9 CHAPTER TWO 2 Literature Review Despite growing recognition of the importance of family in health care and progress in family theory development over the last two decades, transfer of family theory to acute care nursing practice has been limited (Segaric & Hall, 2005). Indeed, in many areas of nursing practice family nursing continues to be what Friedman, Bowden and Jones (2003) referred to as a "stated ideal rather than a prevailing practice" (p. 42). Central to this problem is the gap that exists between family theory and nursing practice (Duhamel, 1995; Friedman, Bowden & Jones; Segaric & Hall). A review of empirical family nursing literature suggests several barriers that make incorporation of family-focused care difficult to realize in clinical practice. These issues converge into three prevailing themes that serve as background to the research questions guiding this study. The purpose of this chapter is to provide a synthesis of the challenges and gaps associated with application of family nursing theory to practice. The first theme addresses issues and problems related to language and conceptualizations of family (Gilliss, 1991; Hutchfield, 1999; Robinson, 1995b) \ The second theme attends to the inadequate explanation of contextual factors affecting family nursing, specifically in acute care hospital settings (Dunleavy, Shamian & Thomson, 2003; Galvin, Boyers, Schwartz, Jones, Mooney & Warwick, 2000; Levine & Zuckerman, 2000; Wright & Leahey, 2000) \ The third theme examines the lack of understanding about interactive Footnote: Some of the critical analysis presented in this chapter overlaps with this article. Segaric & Hall (2005). The family theory-practice gap: A matter of clarity? Nursing Inquiry, 12(3), 210-218. processes between nurses, patients, and family members in the planning and provision of nursing care (Bell, 1995; Hayes, 1997; Levine & Zuckerman; Ward-Griffin & McKeever, 2000) 2.1 Issues Related to Language Several issues are identified here that pertain to language. These include: lack of a common language, the theory-practice gap, family defined as context, family defined as unit, family defined as system, and the need for conceptual clarity. A more detailed discussion of each issue now follows. 2.1.1 Lack of a common language. Well over a decade ago, Gilliss (1991) identified a critical need for a common nomenclature related to the units of assessment, intervention, and evaluation of family across nursing theory, research, and practice domains. In response, various theories and conceptualizations of family began to emerge; however, implementation of family theory to practice 'lagged' behind (Duhamel, 1995, p. 7). Many authors claim that the absence of family nursing practice in areas other than maternity care and pediatrics is due to the lack of a distinct and consistent vision of what constitutes family (Ganong, 1995; Hayes, 1997; Hutchfield, 1999; Robinson, 1995a; Vaughan-Cole, 1998). Moreover, Robinson argued that the variety of definitions of family currently in use do not serve to unify our perspective because a common language that crosses the domains of research, theory, and practice remains absent. To complicate matters, no original family nursing theory exists (Friedman, Bowden & Jones, 2003; Vaughan-Cole, 1998). Consequently, the nature of family nursing is conceptualized using groups of concepts, principles, assumptions and hypotheses borrowed from other disciplines and compiled from existing general nursing theories and adapted to fit various philosophies of family nursing (Friedman, Bowden & Jones; Hanson & Kaaikinen, 2001; Wright & Leahey, 2005; 11 Vaughan-Cole). The multiple conceptualizations that result can leave nurses confused and frustrated in their efforts to discuss and to apply family nursing theory in practice. Indeed, the language of family nursing that has evolved lacks clarity and agreement about what family nursing actually encompasses (Denham, 2003; Segaric & Hall, 2005). As Segaric and Hall (2005) argued: "theoretical explanations [of family nursing] that are accessible are frequently characterized by obscure language and conceptual overlap, which makes achieving a common understanding within practice difficult" (p. 213). 2.1.2 Theory to practice gap. Theoretical conceptualizations that lack clarity and fail to communicate the salient characteristics of the nature of family nursing and to guide nursing care are based on pre-existing constructs as opposed to experiential knowledge gained through doing (Bell, 1995; Friedman, Bowden & Jones, 2003; Hayes, 1997; Vaughan-Cole, 1998). This results in theory that lacks relevance and utility in practice, either because nurses are unable to determine meaning and relevance of the theoretical concepts, or the theoretical concepts fail to fit within the realities of practice - a situation referred to as the family nursing theory - practice gap (Hewison & Wildman, 1996; Segaric & Hall, 2005). The lack of relevance and utility of theory for practice is linked to: 1) the persistent influence of the bio-medical model which promotes ritualistic nursing practice (Anham & Johnston, 2000; Hanson, 2005); 2) the absence and dilution of family theory content in nursing curricula (Hanson, 2005); and 3) the failure of family theories and frameworks to account for the increasing complexity associated with both contemporary families and acute care practice settings (Hartrick Doane, 2005). From the language has evolved three extremely complex yet persistent theoretical conceptualizations of family that are commonly used to guide family nursing education, 12 research, and practice. The three dominant conceptualizations are: family as context, family as unit, and family system theory. Although each conceptualization is generally regarded as distinct, theoretical elements of the concepts and sub-concepts are not only defined differently but also have similarity and overlap. Friedman, Bowden, and Jones (2003) acknowledged that multiple definitions of family can be found across the family nursing literature and there is little agreement about what family nursing actually encompasses (p.36). Friedman stated: "The way family nursing is practiced depends on how the family nurse conceptualizes the family and works with it." (p.36). Her statement not only suggests lack of agreement as to what family nursing means, but also the extent to which it can be practiced. 2.1.3 Family as context. The term family as context is most commonly linked with the family-centered care practice model in which family is conceptualized in orte of two ways: individual as focus and family as context or family as focus and individual as context (Friedman, Bowden & Jones, 2003; Wright & Leahey, 2005). When the family is the focus the nurse concentrates on family members' experience as caregivers coping with illness in another family member. Conversely, when the individual is the focus nursing care is directed primarily toward the individual patient; however, family involvement in that care is considered central to the individual's overall well-being (Friedman, et al.; Newton, 2000). The way in which family is defined and exactly how family members are or could be 'involved' varies from one definition or description of family- centered care to another - if in fact it is described at all. Anham (1994), for example, described family-centered care as a philosophy that calls for partnerships between parents and professionals that support parents in their central caring roles. Newton (2000) emphasized family as an essential part of the child's care and illness experience in her definition of family-centered care. Although each view implies a different view of family (parent(s) to a non-specific description of family), descriptive elements of family-centered care appear similar. These include notions of partnership, collaboration, participation, and communication. Notwithstanding, application of these elements in practice remains problematic and the meaning of each continues to be heavily debated (Gedaly-Duff & Heims, 2001). Issues involved in developing partnerships or collaboration include role stress, overlapping roles, negotiation failure (secondary to opposing care and treatment views), and power struggles between nurses and family members (Levine & Zuckerman, 2000; Newton; Ward-Griffin & McKeever, 2000). While some authors (as in the examples above) consider the focus of care in family-centered care to be the development of collaborative partnerships with family for the purpose of promoting the well-being of the individual patient, others view family care as more holistic in nature. According to Anham and Johnson (2000), the holistic approach shifts the focus of care from the individual to all family members. Emphasis is placed on supporting family functioning as well as recognizing family diversity, vulnerability, risks, and strengths for the purpose of providing individualized care for all family members (Anham & Johnson; Hutchfield, 1999; Newton, 2000). Thus, each individual family member becomes the focus of care, although the scope and nature of the care provided remains unclear. 2.1.4 Family as unit. Confusion about the scope of family nursing and the unit of care in family nursing is further compounded by views of family as client (Friedman, Bowden & Jones, 2003), family as unit (Wright & Leahey, 1990), and family group (Robinson, 1995a). According to Friedman et al. (2003), when family is considered client the entire family becomes the focus of assessment and 14 care. Friedman et al. stated: "The family is now in the foreground, while the individual family member is the background or context. The family is viewed as an interactional system" (p. 37). This description might be interpreted as an attempt to combine two somewhat contradictory theoretical conceptualizations of family (family as focus with individual as context versus focus on the whole family); each view is driven by opposing philosophical perspectives related to parts versus wholes. In practical terms, how does the entire family become the focus of care when individuals are relegated to the background? How can one be either a family member or an individual? Moreover, how might this view of family be implemented in acute care practice settings where the care priorities must first and foremost focus on the ill family member (patient)? 2.1.5 Family system. Wright and Leahey (2005) have viewed the family as a unit or whole, requiring that the focus of nursing care be directed towards both the individual and the family simultaneously. According to Wright and Leahey (2005) when families are viewed as systems "it allows us to view the family as a unit and thus focus on observing the interaction among family members rather than studying family members individually" (p. 32). These authors remind the reader that each individual family member is both a subsystem of the family system and a system in their own right. This conceptualization of family falls under the category of systems nursing, whereby the whole (the family unit) is greater than the sum of its parts (Wright & Leahey). Wright and Leahey further posited that families are best conceptualized as a hierarchy of systems composed of a set or complex of interacting elements. A change in one system causes a change in other systems (Vaughan-Cdle, 1998; Wright & Leahey). Since emphasis is placed on the interaction 15 between and among family members, family nursing from this perspective should include all family members. Although the family systems approach has been identified as one of the most influential of family theory frameworks (Hanson & Kaakinen, 1998), the idea of including all family members in care raises several semantic and organizational questions, especially in acute care practice settings. Definitions of family associated with systems theory, for example, tend to be comprehensive in nature, meaning that families are self-defined (Bell, 1995b; Friedman, et al., 2003; Leahey, Harper-Jaques, Stout & Levac, 1995; Wright & Leahey, 2005). While inviting the individual to define who constitutes the family may provide important access to beliefs about family membership and roles (Bell, 1995b), some important questions include: Whose view of family takes priority? What personal beliefs, practices and/or administrative factors promote or restrict nurses' abilities or willingness to honor families' unique definitions and configurations? If families are considered a hierarchy of interacting systems, then who is responsible for deciding the order of the hierarchy and on what basis? Robinson (1995a) argued that conceptualization of family as parts versus wholes results in confusion related to the focus of care because "persons are conceptualized only at the level of family members, are encompassed by family, and are lower on the hierarchical order" (p. 23). This view of family sets up an artificial separation between individuals and families that Robinson (1995b) referred to as a dichotomy of either/or positions characterized by competition and exclusion. It is essentially reductionistic. Consequently, Robinson (1995a) preferred the term group to describe family because it is inclusive of both family and persons (individual and family member). 16 The family group is defined as "family is foreground and the individual/family members and their relationships are background" (Robinson, 1995a, p. 28). In other words, focus is concentrated on the overall attributes of the family group, while influence of individuals or relationships is obscured. While conceptualizing the family as group resolves the dichotomy between individual systems and family systems, the framework proposed by Robinson (1995a, b), which consists of different views or levels of nursing inclusive of both family and individual systems is, by her own admission, extremely complex and confusing. Moreover, some questions that need to be addressed include 1. What nursing practice implications are associated with shifts from individual systems to sub-systems to family systems? 2. How does a nurse make such shifts in family focus? 2.1.6 The need for conceptual clarity. The persistent lack of conceptual clarity and consensus in terms of what constitutes family and family nursing has implications for future family nursing research aimed at bridging theoretical conceptualizations and the realities of clinical practice. As emphasized by Chinn and Kramer (1999), clarity in conceptual meaning is fundamental to the process of theory development and testing. One of the implications of conceptual clarity is that a similar empirical reality for the concept comes to mind as nurses read the theory (Chinn & Kramer). Hardy (1978) argued that it is important that a theory's significant concepts and conditions be applicable to practice realities and be modifiable depending on the clinical situation. To date, the absence of research linking the various conceptualizations of family with clinical practice is startling (Friedman et al., 2003). Indeed, Hayes (1997) argued that nursing has failed to capture the interconnected nature of family theory and practice. In other words, although family nursing is known to exist in daily practice, Hayes maintained that nurse scholars, 17 educators, researchers, theorists, and practitioners alike have just scratched the surface in their effort to define what family nursing is and determine how to conceptualize its processes let alone explain how to do it. Bell (1995) argued that a search for a common language to describe family nursing must be taken to the practice domain to achieve conceptual description that accounts for the bi-directional influence of what constitutes a family intervention in practice (Bell, 1995). Most recently, Hartrick Doane and Varcoe (2005a; 2005 b) argued that family nursing theory, which was founded on primarily objective forms of knowledge, constrains nurses' ability to respond to the unique health and healing processes of families and limits choices in clinical decision-making. Moreover, Hartick Doane and Varcoe (2005a) maintained that when nursing practice is informed by "multiple knowledges [sic]," including experiential, contextual, spiritual, theoretical, biomedical, ethical, and ideological knowledge, nurses are able to be much more responsive to the unique needs of individuals and families (Hartrick Doane & Varcoe, 2005a). Consequently, Hartrick Doane and Varcoe (2005a; 2005b) advocated for a relational approach to family nursing knowledge development and practice. According to Hartrick Doan and Varcoe (2005a) the relational lens provides a view of the world that is concerned with the interconnections between people, situations, contexts, environments, and processes. In her description of a relational stance, Tapp (2000) emphasized the need for nurses to maintain a respectful regard for families' perspectives in care. O'Sullivan Burchard (2005) proposed that regarding family nursing as a relational practice could advance a shared perspective or what she referred to as an 'ethos of family nursing'. An ethos of family nursing is represented by complimentary interpersonal relations between nurses and families in which nurses respectfully learn about how families make sense of illness and purposefully 18 acknowledge the individuality of families in care (O'Sullivan Burchard; Tapp). Common to these perspectives is the belief that, whether or not a relational approach has the potential to provide a new paradigm of care for families (O'Sullivan Burchard), it can advance family nursing practice because it grounds knowledge development in 'real' experiences and practices that take place in the contingent and ever-changing world of family nursing (Hartrick Doane & Varcoe; O'Sullivan Burchard). Because the value placed on knowledge largely depends on its utility in practice, initiatives in family theory development by practitioners or theories that account for these various perspectives in practice are gaining support (Bell, 1995; Hartrick Doane & Varcoe, 2005a; Hutchfield, 1999). As discussed previously, one of the main objectives of this study is to generate theory to describe how nurses, patients, and family member(s) manage relationships around planning and providing care in acute care hospital settings - and to do so from the perspective of nurses, patients, and families. This work is timely in that it has the potential to make a valuable contribution the development of a conceptualization of family nursing that is meaningful to nurses, patients, and families and aids in efforts to bridge the gap between family nursing theory and acute care hospital practice. 2.2 Issues Related to Contextual Features Several issues related to contextual features in the practice environment that constrain or prevent the application of family nursing theory to practice are now described. Such issues include: failure to account for the work environment, lack of nursing time, and the implications of time for nurse-family relationships. 19 2.2.1 Failure to account for the work environment. A second theme associated with the problem of the family theory-practice gap is the claim that family theory may not adequately account for contextual factors that can constrain nurses' efforts to include families in care. This may be because, to date, relatively limited attention has been paid to nurses' work environments (McGillis-Hall & Kiesners, 2005). Indeed, only in recent years has the true impact of the hospital restructuring of the 1990s been realized. Publications have warned about the serious and rapid deterioration of the quality of work life for nurses and its effects on nurses, the system, patients, and families (McGillis-Hall & Kiesners). In an effort to better understand the impact of the working environment on the health of the nursing workforce in Canada and to make recommendations for improvements, Baumann, O'Brien-Pallas, Armstrong-Stassen, and colleagues (2001) conducted an analysis of published and unpublished literature concerning the health and well-being of Canadian nurses in the workplace and focus group data from nurses and other stakeholders. Based on their findings, they cautioned that, while nurses' work has always been demanding, nurses are facing serious challenges in increasingly difficult work environments. Heavy workloads, long hours, low professional status, difficult relations in the workplace, difficultly carrying out professional roles, and a variety of workplace hazards create stressful working conditions for nurses (Baumann, O'Brien-Pallas, Armstrong-Stassen, et al.), which lead to burnout, decreased nursing recruitment and retention, and negative outcomes for patients, systems, and families (Baumann, O'Brien-Pallas, Armstrong-Stassen, et al.; McGillis-Hall & Keisners, 2005). Similar findings have been reported in follow-up studies by Greenglass and Burke (2002) who found that nursing burnout, particularly emotional exhaustion, and cynicism was linked to stressors associated with hospital 20 restmcturing initiatives, workload, and deteriorating hospital facilities and services (for example, cleanliness, building repairs and general appearance). With the aim of assisting hospitals to address work life issues for nurses and to create quality work environments, McGillis-Hall and Keisners (2005) interviewed Canadian nurses working on medical and surgical units across Canada to determine what work environment issues were important to them. Three main sources of work-related stress were identified: high levels of patient acuity; high levels of workload and understaffing; and lack of adequate patient care. The authors linked increased patient acuity (sicker patients with complex, multi-system conditions) to an aging population and a reduction in moderately ill patients who are being treated on an out-patient basis. Increased acuity was found to increase responsibility, time pressures, and stress in nurses' everyday work. McGillis-Hall and Keisners indicated that escalating workload compounded by staff shortages resulted in nurses reporting a decrease in the provision of quality care and concern for their own and their patients' safety. Although this study clearly suggested that deteriorating nursing work environments affected nurses' abilities to provide quality care, how these working conditions affected nurses', patients', and family members' relationships in acute care hospital settings remained unclear. Several authors have maintained that nursing shortages, decreased length of patient stay and increased patient acuity have resulted in less nursing time spent with patients and families (Galvin, Boyers, Schwartz, Jones, Mooney & Warwick, 2000; Levine & Zuckerman, 2000, Wright & Leahey, 1999; 2000). Authors claim that sicker patients require closer supervision and complex nursing interventions leave little or no time to meet emotional or other needs of patients or families (Dunleavy, Shamian, Thomson, 2003; Friedman, Bowden & Jones, 2003). Rutledge, 21 Donaldson and Pravikoff (2000) argued that family needs are often inadvertently neglected by staff focused on moment-to-moment demands of patient care. 2.2.2 Lack of nursing time. Dunleavy et al. (2003) examined the effects of hospital organization and staffing on patient outcomes and identified the problem of insufficient nursing time. The over eight thousand (8,000) nurses they surveyed reported that it was a struggle to keep up with increasing numbers of non-nursing roles and responsibilities (for example, paperwork, housekeeping, switchboard operator); they felt burnt out and dissatisfied with their work. They regretted their lack of time to provide balance between medical and psychosocial care for patients and their families (Dunleavy et al., p. 25). Although lack of nursing time is often cited in the literature as a primary barrier to applying family theory in acute care hospital settings, and we would expect that inadequate time for patients' needs would translate into lack of time for family needs, there is limited evidence to support this conclusion. Indeed, the publications reviewed by this author revealed a dearth of empirical research which specifically investigated how time promoted or inhibited family nursing practice in acute care settings. In a pilot study examining family members' experiences of their role in a hospital (n=70), Astedt-Kurki, Paunonen and Lehti (1997) found that only one-third of the family members sampled felt that nursing staff were seriously interested in the family's well-being. Participant responses were partly based on the lack of spontaneous information they felt they received from nurses, as well as the effort required on their part to seek out nurses for information and support. Although the authors suggested that decreased nursing time might be a factor in the lack of communication and information shared between nurses and 22 families, the researchers did not account for contextual explanations from either families' or nurses' perspectives. Keatinge, et al. (2002) noted that consumer involvement (broadly defined as consumers of nursing care services) in health care is desirable but generally not implemented due to poor consensus about what participation means and nursing apprehension about partnering with families due to lack of time and their extra space requirements. In an effort to determine what barriers exist in nurse-consumer partnerships and why, Keatinge et al. conducted an 8 month pilot study using an audio-taped workshop format. Findings indicated that communication was the principal barrier identified by both acute care nurses (N = 199 Registered Nurses) and consumers (N = 36). The contexts in which communication barriers occurred were grouped by theme. They included: information transfer, documentation, education, personal and interpersonal situations, guidelines and policy, organizational structures, politics, and, lastly, resources and time. Each theme was labeled on the basis of recurring terms used by participants to describe a communication-related strategy or situation. Despite the fact that several systemic themes, including resources and time, were identified, the study was limited by the lack of consideration of possible relationships between themes. Moreover, the second phase of the study - which was to generate strategies to improve communication, focused on only the information transfer, documentation, and education themes, as these were the top three priorities identified by participants. Consequently, recommendations arising from the study were limited to improving communication processes in isolation of the effects of context on communication processes. Using a descriptive survey design, Galvin et al. (2000) sampled 193 participants (either maternity patients or parents of a hospitalized child) in a large tertiary care hospital to explore participants' perceptions related to the family-centered care philosophy being piloted. One of the 23 hypotheses of the study was that, despite widespread support of family-centered care, family-centered philosophy remains inconsistently operationalized in hospitals due to numerous barriers to implementation including space and time constraints and the ability to meet the physical needs of the patient. Results suggested that, faced with limited time, nurses' priorities may be the patient's physiologic needs and the technological demands of the environment. They did not seem to perceive their time and resources to be adequate to provide support to patients and their families (Galvin et al.). Although the authors found that items measuring staff inclusion of family in care were rated higher on level of importance and lower on level of occurrence, they did not explain this discrepancy or its effects on nurse-family relationships. 2.2.3 Implications of time for nurse-family relationships. Limitation in nursing time has also been associated with the type of nurse-family relationship that is established in the clinical setting (Levine & Zuckerman, 2000). Although most family nursing theoretical frameworks call for the development of collaborative partnerships between nurses and family, time constraints may force nurses to assume a less collaborative and more functional approach with families. Hutchfield (1999) referred to this as the role of gatekeeper and dominant player in the partnership where nurses decide in which care families can participate and delegate accordingly. Nurses' abilities or willingness to invest time in the establishment of collaborative partnerships with families has also been linked to nurses' beliefs about their roles and responsibilities to patients versus families. As Levine and Zuckerman (2000) pointed out: "there are deeply rooted reasons for conventional attitudes that overlook or denigrate the role of family members in patient care and decision-making" (p. 6). Medical practice and bioethics, which have both heavily influenced the development of nursing practice, have viewed individual patients as 24 the primary if not sole focus of concern and care. This traditional view of patient continues to direct medical and nursing allocation of time in many settings, thereby placing family in what Yates (1998) referred to as an ambiguous position in health care. The individual focus dominating health care limits opportunities for professionals and families to understand each other's issues and concerns (Yates). Wright and Leahey (2005) supported the notion that social and contextual factors play an important role in the distribution of nursing time. According to these authors, social coordination is a function of time that gives it instrumental value in health care. In nursing, time is "socially and culturally coordinated, highly ritualized and therefore honored" (Wright & Leahey, p. 264). The degree to which families are also honored may well determine the likelihood that nurses will alter their practice contexts to increase their caring for and inclusion of families in practice (Wright & Leahey). Despite evidence that families can have significant influence on promoting the well-being of their ill family members, families continue to be widely excluded or marginalized by health care professionals generally and by nurses specifically (Astedt-Kurki, Paunonen & Lehti, 1997; Wright & Leahey, 2005). Such an observation begs the question: What does it mean to honor families in health care so that they are seen to be worthy of nursing practice time? Wright and Leahey maintained that nurses must alter their constraining beliefs about involving families; however, it is acknowledged that altering practice is complex because it involves a combination of administrative support, family-friendly facilities and nurses who are committed, knowledgeable, and skilled at involving families in care (Wright & Leahey). Wright and Leahey (2005) also emphasized that any involvement of family members is superior to no involvement. The authors introduced the 15-minute assessment framework consisting of five essential ingredients: manners, therapeutic conversations, family genogram and ecomap, therapeutic questions, and commendations to account for limited nursing time. The framework was designed to provide an efficient means of incorporating family interviews into nursing care. Wright and Leahey argued that, for nurses to make time in their hectic schedules, even for a brief interaction, they should modify their beliefs to embrace the philosophy that illness is a family affair (Wright & Leahey). But is it simply a matter of changing nursing beliefs or are other larger forces at work creating barriers to nurses' best intentions to involve families? What are the structural factors that reinforce socially and culturally coordinated nursing time in ways that exclude families? How much is actually understood about the nature of constraints such as lack of time and other barriers that are only surfacing as a result of the challenges in contemporary practice settings? 2.2.4 Other contextual issues affecting nurse-family interaction. In addition to time constraints, legal, administrative, and building design issues add complexity to providing family care and require further investigation in terms of their effects on nurse - patient and family relationships. Levine and Zuckerman (2000) argued that the structure and practices of the health care system force health care professionals into adversarial positions with families. According to these authors, while some clinicians are sensitive to the needs of families, many have their efforts to engage and support families thwarted by organizational requirements and poor reimbursement. For others, the dominance of medical and bioethical models that privilege professional knowledge and some patient needs create situations in which families are considered as suspect and distracting entities that disrupt institutional routines and authority (Levine & Zuckerman). These authors claim practioners may also fear litigation from angry or disgruntled family members which may prompt them to focus more on individual 26 patient needs and limit or prevent family members from observing or participating in direct care in case something goes wrong. Institutional policies that limit the number of people that can visit safely and/or any other policies that restrict family members' access to their ill family members are also practice realities commonly believed to decrease family nursing initiatives. In a study conducted by Galvin et al. (2000) on maternal/child units, parents surveyed rated respect, collaboration, and support as critical elements of a family-centered care philosophy. Respect included having access to their child, making them feel more welcome in the hospital environment, assisting them in maintaining their parental role, and acknowledging parents' rights to question medical recommendations. As a result of this study, visitation policies in the participating hospital were expanded to include grandparents and siblings, the frequency that parents were required by policy to leave the bedside (e.g. during nursing report) was decreased, and greater overall emphasis was placed on enhancing and supporting nurses' efforts to value parent participation in care. The outcomes of these changes were not reported. Barriers to family access associated with building design are cited in several studies; however, the importance of these issues varies. In their synthesis of the literature, Rutledge, Donaldson, and Pravikoff (2000) found that implementation of family-centered care is often inhibited by structural or architectural barriers, such as waiting rooms that isolate family members and/or prevent them from staying at the bedside. Similarly, Lee and Craft-Rosenberg (2002) noted that formal and informal structures associated with hospitals, as well as inconsistencies in visiting policies, posed real barriers to the provision of services and participation by families in care. Verhaeghe, Defloor, Van Zuuren, Duijnstee, and Grypdonck's (2005) review that pertained to the needs and experiences of family members of intensive care 27 patients indicated that the 'practical needs' of families in the hospital environment are generally considered less important by hospital staff and family members. These authors cautioned that the notion of physical needs can be interpreted to include everything from visiting hours to the availability of material items and conveniences such as comfortable firrniture, blankets, a place to lie down, and close proximity to a bathroom. Generally, family members placed higher importance on the need for flexible visitation rights than did hospital staff. Material needs tended to be ranked lower in priority by both hospital administrators and family members. In their analysis of the perceived needs of Jordanian families of hospitalized, critically ill patients, Al-Hassan and Hweidi (2004) found that families' needs for information ranked highest in priority, whereas items associated with comfort including visiting times, waiting rooms, and other comfort measures were ranked lower. Incorporating family nursing care in acute care settings is complex and involves interacting and possibly opposing contextual and systemic forces. While nurses are being called upon to involve families in care, time constraints require that they develop ways to do so in an expedient and effective way (Friedman, Bowden & Jones, 2003; Wright & Leahey, 2005). It is clear that the degree of family-centeredness that is attainable may be dependent, at least to some extent, on any one of a number of factors including the set of beliefs and values held by the nurse (Wright & Leahey), the philosophy or support provided by the system within which the nurse works (Friedman et al.), and physical barriers that prevent or discourage family involvement in care in acute care hospital settings (Rutledge et al. 2000). Despite many studies that address contextual and systemic factors associated with nurses' abilities to incorporate families into their practice, there seems to be limited agreement and understanding about the significance of these factors and the extent to which they promote or inhibit nursing efforts to involve families in patients' care. Of particular note is the lack of research that takes into account the perspectives of nurses, families, and patients. For family nursing to advance in acute care settings, we must gain a better understanding of the contextual and systemic factors or structural conditions that affect nurse, patient, and family relationships. 2.3 Issues Related to Interactive Processes A third theme associated with the family theory - practice gap concerns the paradox of the importance placed on nurses' work with families and the lack of understanding about the nature and development of nurse-family relationships in acute care settings. While theoretical knowledge and conceptualizations of family have grown and evolved at an exponential rate, understanding of processes that occur between nurses and families in care has been slow to develop (Bell, 1995; Vaughan-Cole, 1998; Griffin-Ward & McKeever, 2000). The advancement of theory rather than practice has been linked to gaps and limitations in knowledge about how nurse-family interactions are conceptualized or actualized in practice. Some scholars, for example, have suggested that family nursing theory based solely on objective knowledge fails to promote nurses' development of interactions with families beyond a superficial level, prompting them to advocate for a more relational conceptualization of family (Hartrick Doane & Varcoe, 2005; O'Sullivan Burchard, 2005; Wright & Bell, 2004). Hartrick Doane and Varcoe have maintained a relational stance is necessary, because family theories are decontextualized and depersonalized and, therefore, lack meaning and utility in practice. Several other problems hamper the conceptualization and development of nurse-family interactions in clinical practice. For example, the literature privileges research for theory development over research examining the interface between family theory and practice (Hayes, 1997; Vaughan-Cole, 1998). Hayes identified problems with the lack of resources available to assist practicing nurses to understand the possibilities, expectations, and scope of their interactions relative to families. Furthermore, Friedman et al. cited lack of evidence to support the efficacy of family nursing strategies and programs. In the context of cancer patients and their families, Yates (1999) argued that conceptualizing the relational and reciprocal nature of family members' needs and determining ways of dealing with conflicting needs in practice are challenging. Allen (2000) argued that how family caregivers interact with the 'service system' in the acute care hospital context is poorly understood. Finally, Wright and Bell (2004) maintained that only by altering their way of thinking about who is patient will hospital nurses gain greater insight into families' experiences of illness and discover intervention alternatives. Many studies have presented ideal goals for nurse-family relationships, but have based their claims on the perspectives of either nurses or families, but not both groups of participants. Few studies have sought to understand multiple perspectives, and, no studies have examined family nursing interactions or interventions from nurses', patients' and families' perspectives. For example, a frequently cited grounded theory study by Robinson and Wright (1995) identified giving of information, maintaining a non-judgmental, compassionate, and genuinely interested stance, and the use of therapeutic conversations as nursing interventions considered most helpful by families. The study was limited by its location in the family nursing unit -a research and educational unit associated with the University of Calgary, its focus on families coping with chronic illness, its inclusion of only Caucasian families, and its lack of input from nurses about practice realities that can impede or prevent implementation of such strategies. The recommendations for nursing from the study were also limited because nurses' views were not included in the sample. 30 Astedt-Kurki, Lehti, Paunonen, and Paavilainen (1999) explored experiences of family members when their next of kin were admitted to hospital. The authors identified three main classifications of family experience: fear and worry; disruptions and changes in everyday life; and extended periods of time spent helping the patient. They recommended that nursing strategies, aimed at supporting and assisting families in providing care, ought to take into account the uniqueness of families' experiences of the hospitalization of a relative. Although the study provided insight into the experiences of families in their roles as caregivers, it is limited by the fact that 80% of the respondents were female spouses of the patients and patients were receiving treatment at an acute neurological ward of a large hospital. Because data were collected via questionnaires mailed to a family member who patients identified, neither patients' nor nurses' views were investigated. A consistent lack of nursing input for studies about relationships with nurses could contribute to nurses' feelings of guilt, anger, and confusion associated with the disparity between how nurses ought to 'be with families' and factors that prevent them from doing so -even if they believe it is important. Nurses may also feel betrayed due to the lack of investigation into how they promote family involvement. Practicing nurses may view studies that do not include their perspectives as irrelevant, because they fail to resonate with the realities of practice experienced by those working in the clinical setting. Nurse-family relationships are also poorly understood in terms of contextual factors, including culture and long held beliefs associated with families and family roles in health care. Lee, Chien, and Mackenzie (2000) noted the effect of critical illness on family members and the importance of nursing care inclusive of family is well documented in American studies. Unfortunately, these studies are primarily based on the Western culture and fail to account for 31 differences across various ethnic or racial groups. Indeed, Lee et al. expressed surprise at the lack of research related to the Chinese population given their cultural beliefs about the significance of family and the importance of maintaining family relationships, roles, and obligations in order to preserve equilibrium within the family system. Levine and Zuckerman (2000) noted the potential for increased conflict between families and health care providers when families operate from different religious, cultural, or ethnic backgrounds than those of care providers. Beliefs and expectations that nurses' hold about families and, conversely, beliefs and expectations that families hold about health care providers require further investigation to determine their effects on promoting or inhibiting nurse-family relationships. Kellett (2000) emphasized the competitive relationships that often occur between families and professional caregivers in the context of long term residential care facilities, because families are traditionally expected to relinquish care to the bureaucracy and failure to do so may be looked upon as interference. Families often work to improve the quality of care experience for their kin; however, some studies have indicated that professional caregivers may view this as interfering with routines (Kellett; Levine & Zuckerman, 2000) or implicit criticism that nurses are not doing a good job (Kellett). Levine and Zuckerman have argued that often family involvement is seen as a challenge to the power and authority of professionals and institutions. Health professionals often consider families to be dysfunctional and incapable of participating in decision-making (Levine & Zuckerman, 2000). Such judgments block the development of collaborative nurse-family relationships. Family behaviors, especially during times of stress or crisis, are often misinterpreted or poorly understood by healthcare providers (Kellet, 2000). Levine and Zuckerman suggested that many health professionals believe that families are generally disintegrating and dysfunctional. Such beliefs have been perpetuated by a few 32 legendary, truly dysfunctional family cases, by families that are known in terms of legal precedent, or by heroic families by which all other families are measured. Family nursing research that aims to explicate underlying beliefs and assumptions held by families, patients, and nurses in the context of acute illness episodes will increase our understanding of the nature of their relationships under common conditions. Notwithstanding the various constraints and challenges in explicating nurse-family interactions in clinical practice, a number of factors - as perceived by nurses, patients, and family members - that facilitate positive nurse-family relationships appear repeatedly in the family nursing literature. In a complex, multi-phased grounded theory study to examine the development of nurse-family relationships in the intensive care unit (ICU), Hupcey (1998) identified numerous facilitative and constraining strategies used by nurses and family members. Due to the extensive nature of the study and the number of outcomes reported, the findings are presented here in an abbreviated format. Hupcey (1998) indicated that strategies identified by nurses as facilitative in demonstrating their commitment to families included: spending time, providing explanations, encouraging participation, and anticipating family members' needs. Those facilitative in building relationships with families were spending time, getting to know family members, and sharing personal information. Nurses used strategies to demonstrate ongoing involvement with families, such as spending increased time, becoming a family advocate, and stretching or breaking the rules. Strategies used by family members to develop relationships with nurses included assessing for evidence of nursing competence, kindness, and genuine interest toward the ill family member; making an effort to be 'good visitors'; and displaying trust by accepting nurses' explanations, seeking nurses' advice, relinquishing vigilance with the patient. Being over-or 33 under- involved in patient care and displaying a lack of trust in the nurse were family behaviors reported by nurses that they perceived inhibited the development of nurse-family relationships (Hupcey). In Hupcey's (1998) study, nursing behaviors perceived by family members to inhibit nurse-family relationships included depersonalizing the patient by not calling the patient by name and not making eye contact; not encouraging family participation in care; maintaining an efficient attitude by acting too busy to answer questions; displaying a lack of trust in the family by not listening to their concerns, and asking them to leave the room. Upon verifying the occurrence and negative impact of their behaviors, nurse participants cited being preoccupied, extra busy, and feeling the need to retain power and control as contributing factors. Although limited to ICU situations and not inclusive of patients' views, a major conclusion of the study was that most families believed that facilitating relationships with nurses was of benefit the patient. Conversely, they viewed alienating the nurse as negatively affecting many aspects of patients' care including obtaining information and participating in decision-making related to the patient -a reminder that nurses must be mindful of their actions and their effects on families (Hupcey). Another limitation of this study was that it did not examine strategies that facilitated or constrained the development of nurse-family relationships from nurses and family members' perceptions about self and others, and structural features. Although limited to the perspective of nurses, Astedt-Kurki, Paavilainen, Tammentie, and Paunonen-Ilmonen (2001) found that the most important factors that facilitated nurse-family interactions in acute care hospital settings from the perspective of staff nurses included discussion opportunities provided by staff and the patient's positive attitude toward staff-family interactions. Additional factors that contributed to the development of nurse-family interactions 34 included the family's level of interest in the patient and the confidence that family members showed toward the nurse. Factors that were considered by nursing staff to complicate nurse-family interactions included: nurses' haste; shyness by either the nurse or especially the family in terms of approaching each other; inaccessibility of staff and shift work; and patient's resistance to nurse-family interactions. Also, despite nurse participants' perceptions of their interactions with family as being fairly important, interactive episodes were considered rare, were seldom nurse-initiated, and almost always were limited to aspects of the patient's condition and care. Contrary to Hupcey's (1998) results, although nurses' haste was also seen as a primary complicating factor in nurse-family interactions, it was the nurses' attitudes toward family members and their recognition of the family's importance to the well-being of the patient that was seen to underlie hasty behavior by nursing staff as opposed to lack of time (Astedt-Kurki et al.). Nurses' attitudes towards patients' family members and the importance of family in the well-being of the patient, as well as the value of providing information to family members, have clearly emerged as dominant themes in the literature associated with interactive processes between nurses, patients, and families (Astedt-Kurki et al., 2001; Auerbach et al., 2005; Majassaari, Sarajarvi, Koskinen, Autere & Paavilainen, 2005; Wright & Bell, 2004). In a study measuring family members' satisfaction with needs being met during intensive care hospitalization, Auerbach et al. found that, although family members experienced high levels of anxiety and fear associated with sudden hospitalizations of ill family members, their needs for information, clear explanations, and knowledge about equipment being used, especially upon admission, were least satisfied. Nursing initiatives to improve communication (for example, family conferences) foster a sense of optimism and control in family members by maintaining a 35 positive attitude and encourage participation in care. Those elements were linked with a reduction in family member anxiety (Auerbach et al.). In Majassaari et al.'s (2005) survey of day surgery patients about the importance of information and emotional support provided to their family members, the need for emotional support to ease family members' nervousness, fear, and anxiety associated with the surgical process was noted. Patients also identified strengthening of patient-family relationships around surgical episodes and enhancing family coping and well-being through family members' presence and mutual support. Because patients perceived that nurses were too busy, did not meet with family members, and provided inadequate explanations or no information at all to families, recommendations from the study call upon nurses to be "more sensitive, perceptive, and present" to the unique needs of patients and families (Majassaari et al.). Despite the dearth of family nursing research aimed at identifying strategies for nurse-family interaction and limitations in studies that have been undertaken, a common theme underscores the importance of ways of developing relationships with families and incorporating them into nursing care. The assumption is that illness in one family member not only affects all other family members, but also affects them in ways that are unique to the patient and the family (Tapp, 2000b; Wright & Bell 2004: Wright & Leahey, 2005). Seeking to establish nurse-family relationships for the purpose of understanding and meeting the unique needs of families may be the primary goal of family nursing, but it is far more complex than meeting needs. As discussed in this section, numerous facilitative and constraining strategies, behaviors, and situations have been identified and others are yet to be explicated. Examining how personal factors (beliefs, values, experiences) affect nurses', patients' and family members' perceptions of their interactions to plan and provide patient care would contribute to the identification of nursing interventions with families that are both helpful and realistic within the constraints of the acute care hospital settings. 2.4 Summary The centrality of families to health care continues to gain support; however, there are challenges to efforts to incorporate family theory into acute care practice. Based on a review of extant family literature, it is evident that there are issues, constraints, and barriers that make family nursing difficult to realize in clinical practice. These difficulties converge under what has been described in this chapter as three prevailing themes: issues related to language, contextual features, and nurse-family interactive processes. They serve as the basis of the theory-practice gap observed in family nursing. I have indicated that there is progress being made toward identifying strategies that facilitate the development of nurse-family relationships in practice. Examples of facilitative themes included: providing information, being supportive, maintaining a positive attitude, and spending time. Limitations in the knowledge obtained to date and how it has been generated have prompted a call for an examination of practice in order to understand how nurses, families, and patients conceptualize each other and navigate or remove barriers to engage with each other in care. In the next chapter, I present the theoretical perspective of symbolic interactionism selected to guide this study. 37 CHAPTER THREE 3 Theoretical Framework: Symbolic Interactionism The purpose of this chapter is to describe the theoretical framework of this interpretive research study. Interpretivist research is concerned with the meaning of human social action. Finding meaning in the action of others requires interpretation (Schwandt, 2000) or what Charon (1979) referred to as 'perspective'. A theoretical framework reflects a researcher's theoretical perspective or a particular set of epistemological, ontological and methodological assumptions and beliefs about the world and how it is understood and studied (Denzin & Lincoln, 2000) and provides justification, context and logical grounding for the particular research design implemented (Crotty, 1998). In other words, one's theoretical perspective allows one to make sense of and describe the reality to which one is exposed. There are a number of theoretical perspectives within the interpretivist paradigm that espouse philosophical views about the nature of reality (ontology) and of knowledge (epistemology). Each perspective makes particular demands of the researcher in terms of the mode and meaning of the process of inquiry undertaken and the interpretation of findings (methodology). In conducting interpretive research, I must consider not only how I am situated philosophically but must also examine the fit between my beliefs and the theoretical perspective that informs the strategy of inquiry I use. I must also be concerned with justifying and defending the process of inquiry in which I have engaged by making it explicit to the scrutiny of others (Crotty, 1998) and by making explicit the philosophical stance that provides the context for the research process and grounds its logic. Since my research study is concerned with gaining understanding about issues of language, cornmunication, interaction and social context in relation to the management of relationships among patients, nurses, and family members, I have chosen symbolic interactionism to ground my study. Symbolic interactionism focuses on the meaning of basic social interactions from the perceptions, attitudes and values of the actors (Crotty, 1998). To set the stage for my methodology and to make explicit the assumptions upon which this study rests, the purpose of this chapter is to provide a more detailed description of the philosophical and theoretical underpinnings of symbolic interactionism. 3.1 An Overview of Symbolic Interactionism Symbolic interactionism is commonly described as a unique and relatively distinct American sociological and social psychological perspective on life, society, and the world (Blumer, 1969; Crotty, 1998; Denzin, 1992). Although the thoughts of many prominent scholars (e.g. Cooley, James, Dewey, Darwin) over several decades contributed to the intellectual foundation upon which symbolic interactionism rests, the perspective is based primarily on the work of social psychologist and philosopher George Herbert Mead and sociologist Herbert Blumer. While Mead is frequently referred to as the champion of symbolic interactionism, it is Blumer, a long time student of Mead's, who is credited with interpreting and developing Mead's ideas into a more systematic sociological approach. Indeed, it was Blumer who coined the term Symbolic Interactionism in 1937. According to Blumer the label symbolic interactionism refers to "the type of interaction that makes significant use of gestures or symbols" (Blumer & Morrione, (Ed.), 2004, p. 22). Symbolic interaction rests on a process of social interaction based on interpretation of symbols. Inherent to this view of social interaction are, as expressed by Blumer (1969), three basic assumptions or premises. The first premise is that human beings act toward things (gestures or symbols) based on meaning. Secondly, meaning of symbols is derived from or arises out of 39 social interaction. Third, meanings are perceived and modified through an interpretive process. In other words, human beings act on things within their social world based on the meaning things have for them. Meaning is derived from interaction with self and others and in turn influences further interaction (Denzin, 1992). 3.2 Pragmatism In the formulation of the main premises underpinning the social interaction perspective noted above, both Mead and Blumer were strongly influenced by pragmatist philosophy. Pragmatism emerged from the philosophical beliefs of such scholars as Pierce, James, Dewey and Mead - all of whom were disillusioned with the prevailing determinist philosophy (e.g. positivism) because it resulted in views about human behavior that lacked relevance to 'everyday' situations people experience (Atkinson & Housley, 2003; Charon, 1979, Prus, 1996). Pragmatist philosophies emphasize the importance of exploration and interpretation of human engagement in the social and natural world (natural sciences), (Atkinson & Housley). Pragmatists believe that human beings are not "passively shaped" (Charon, p. 30). Mead (1938), for example, maintained that people are emergent beings because of their ability to reason and use symbols. In other words, individuals act as independent forces who are actively involved in planning and carrying out action in response to their environment. They cannot be explained by simply isolating forces that act upon them (Charon). Human action is characterized by continuous adjustment or adaptation in order to master the conditions of their environment (Reynolds & Herman-Kinney, 2003) and to find solutions to practical problems (Prus, 2003). Further to the central pragmatist principle of human beings (organisms) in interaction with their environment is the notion of selective perception. Selective perception is the process by which individuals attend to only certain aspects of a situation (Reynolds & Herman-Kinney, 40 2003, p.256). Human beings are constantly bombarded with stimuli; they must select those stimuli or aspects of the environment that are relevant and/or have utilitarian value to the act in which they are engaged. Through this selection process, individuals actively engage with their environments and are therefore considered self-determining vis-a-vis their environments (Atkinson & Housley, 2003; Crotty, 1998; Reynolds & Herman-Kinney). Pragmatism, as it relates to interpretivism, also focuses on communication stimuli and their consequences (Maines, 1997). Communication is grounded in the human ability to use symbols that are associated with various cultural and social utilitarian meanings that affect social interactions. Blumer (1969) maintained that human beings act toward symbols or objects (everything that the human being may note in his world) based on the meaning they attribute to the objects. In other words, meaning is derived not from the object itself but instead is socially constructed on the basis of how people act in response to the object (Crooks, 2001). Since symbols may have multiple meanings, individuals engage in a process of selecting and communicating/acting on the meaning that is most practical to the situation. Actors are capable of reason, expressing preferences, and exercising choice (Lai, 1995). In the process of studying interactions among nurses, patients, and families it is important to consider key concepts such as human agency, communication, meaning, and social process. These concepts strongly resonate with symbolic interactionism (Prus, 2003). Not surprisingly, the similarities between pragmatism and symbolic interactionism have led many to speculate that it is through the theoretical perspective of symbolic interactionism that pragmatist philosophy entered the world of sociology and social psychology (Crotty, 1998; Reynolds, 2003). Prus (2003) maintained that many of the pragmatist insights about human behavior have been "absorbed by and recast" into the symbolic interactionist perspective of 41 today (p. 48). This intimate relationship between pragmatism and symbolic interactionism becomes even more apparent when the ontological, epistemological and methodological underpinnings of symbolic interactionsim are explored - the focus to which I now turn. 3.3 Ontology Two pragmatist philosophical principles are reflected in the basic assumptions about reality held by symbolic interactionists. These ontological principles state that reality exists only as interpreted and defined by the individual, and secondly, that individuals define objects or symbols on the basis of perceived utility (Charon, 1979). Each principle will now be discussed in turn. The first principle is that reality is only possible through an individual's own intervention or in other words, reality does not exist without our interpreting and defining it (Charon, 1979). This is not to suggest that there is no objective reality outside human interpretation, indeed, symbolic interactionism maintains that a physical objective reality does exist. Physical objects in relation to human beings are considered social objects that represent meaning from the perspective of individuals as they interact with others. According to Blumer (1969): The meaning of a thing for a person grows out of the ways in which other persons act toward the person with regard to the thing. Their actions operate to define the thing for the person. Thus, symbolic interactionism sees meanings as social products, as creations that are formed in and through the defining activities of people as they interact (p. 4-5). Reality is expressed through a system of symbols that reflect social life. Blumer suggested that social interaction through symbols assumes two forms, which Mead designated as 'conversation of gestures' and 'use of the significant symbol' (Blumer & Morrione (Ed.), 2004). A gesture is seen as an abbreviated or portion of a larger action - a stimulus that presupposes a response in the form of a forthcoming action. Because the meaning of gestures is 42 implicit in the gesture itself, response is characterized as instinctual or reactive as opposed to interpretive. Blumer (2004) writes: In the conversation of gestures, the organisms in interaction respond to each other's gestures without identifying the meaning of the gestures. Each organism responds unreflectively to the gesture of the other with an action already organized for release; the response is not preceded by, or based on, an interpretation of the gesture (Blumer and Morrione (Ed.), p. 89). Social interaction on the basis of 'conversation of gestures' therefore, is a process that brings people together in interaction by implicating others and evoking a response. Gestures become symbols when individuals interpret and respond to the gesture on the basis of interpretation (Blumer, 1969; Blumer & Marrione (Ed.), 2004). The distinguishing feature between conversation of gestures and the significant symbol is the process of interpretation. This process requires that the individual identify what a gesture signifies and then devise various kinds of responses to it. The defining characteristic of'significant symbols' is that a symbol is part of a line of action that is picked out and interpreted to point to a larger act of which it is part (Blumer). In other words, the significant symbol belongs in the act. Underlying the act of 'picking out' a line of action is the notion of utility. The second basic principle of pragmatist philosophy that pertains to the ontology of symbolic interactionism is that human beings define objects they encounter according to the perceived usefulness of the object (Charon, 1979; Crotty, 1998). In other words, reality as determined through the interpretive process of deriving and modifying meanings about symbols occurs on the basis of the utility of said meaning for dealing with situations of daily living encountered by the individual (Blumer, 1969). From this principle emerges symbolic interactionist assumptions that the meaning human beings attach to symbols (person, objects, situations) not only determines how they will act toward those things (Annells, 1996; Crotty; Lai, 1995), but also 43 affects the process used to handle and modify meanings in response to the situation encountered (Annells; Crotty). The interpretive process may include reflecting on past socialization and social interaction in relation to an object that reinforces old meanings and patterns of action or reflecting on current or future (imagined) interaction that, in turn, may give rise to new and creative meanings (Annells). Although actors (individual, group) are active agents in denning and acting in response to their social environments, this is a reciprocal relationship in that both the actor and the environment (society) affect and are affected by the other (Charon, 1979). Charon writes: "We share with others a definition of the world and its objects.. .each time we interact with others we come to share a somewhat different view of what we are seeing" (p.54). It is, according to Charon, "this interaction that gives rise to our perception of what is real and how we are to act toward that reality" (p. 54). A key assumption of symbolic interactionism then is that reality can only be understood from this contextual, reciprocal relationship (Fine, 1993). Since objects or symbols are assigned temporal meaning, that is meaning based on individual and social perceptions within past, present, and future contexts, symbolic interactionism also assumes the existence of multiple meanings or interpretations of reality. For the symbolic interactionist, the 'self is a quintessential example of an object encompassing multiple meanings over time. In relation to the social nature and origin of self, Mead (1938) hypothesized that individuals enter into their own experiences as an object in experiential transaction with the social environment. Consequently, the definition of self is affected by one's interpretation of the perspectives of society and social interaction with others in society. One's definition of self is also conceptualized and re-conceptualized through interaction with the "self (Blumer, 1969). Interacting with the 'self means that individuals have the ability to 44 communicate with and analyze the self. The reflective nature of self also allows a person to see him or herself as "object" (Blumer). Charon (1979) states: "The person imaginatively gets outside of his or her person and looks back on self as others do. This process depends to a high degree on taking the role of others, both significant others and reference groups, to see self from their perspective" (p. 79). In determining action or making choices, people determine those things or symbols that have meaning for them in a given situation through a process of reflection and self-communication. By virtue of this process of communication with self, interpretation becomes a matter of handling meanings (Blumer). By creatively grouping various meanings, new meanings are created that ultimately direct action. Obtaining individual perspectives and meanings, therefore, is critical to a research process which seeks to understand and describe how nurses, patients, and family members manage relationships when planning and providing care in acute care settings. The concepts of process and time must also be emphasized in relation to symbolic interactionist's view of reality (Charon, 1979). Symbolic interactionism assumes that all that is associated with being human (the individual, society, the mind, self, and truth) are not things but dynamic processes. Biological, structural, and cultural (e.g. roles, social expectations, norms, values) factors or constraints associated with any situation in which a human actor finds him/herself are taken into account in the course of determining action (Snow, 2001). Consequently, symbolic interactionism may be considered a theory of experience (Denzin 1992). Reality is judged by what is being observed and experienced; it is determined based on interaction with self and others at any given moment in history. While reality is organized in terms of time it is not static. Rather, reality is constantly transformed through a process of interaction as it occurs across time in relation to pasts, presents, and futures (Hall, 1987). 45 3.4 Epistemology Two general pragmatist principles are reflected in symbolic interactionist assumptions related to the nature of knowledge. The first principle described by Charon (1979) is that "knowledge for the human being is based on its usefulness" (p. 29). Human beings learn and remember based on their judgments of how knowledge was useful in a given situation. Humans are thought to be continuously testing hypotheses related to ideas and action in various contexts and knowledge is refined or changed based on the outcome, result, or consequence (Charon). The second pragmatist principle reflected in the epistemology of symbolic interactionism is derived in part from the objectivist orientation that asserts the position that what we know is determined by what we are able to observe (Annells, 1996; Denzin & Lincoln, 1998). For the symbolic interactionists, human or social action is the quintessential reference point to understanding both the individual and the group. As Charon (1979) maintained, it is from empirically observed human action that we are able to understand the human organism. However, symbolic interactionists also maintain what we know is more than what we observe or are able to see. In other words, to understand action one must infer meaning from the action (Charon). Inference of meaning about actors or between actors is generally based on consistency of action over time (Mead, 1938). From the first epistemological principle emphasizing that knowledge is based on its perceived utilitarian value emerges several symbolic interactionist assumptions. First, knowledge is not a static concept. Knowledge is thought to be constructed and re-constructed through a process of social interaction (Crotty, 1998; Donmoyer, 1999). While constructionists maintain that meaning is derived from an interplay between object and subject, and subjectivists argue that meaning is imposed on the object by the subject based on perceptions derived from social 46 interaction, both positions rest on the symbolic interactionist assumptions that knowledge is socially situated (contextual), dynamic, and can only be understood through the perceptions of self and others (Ambert & Alder, 1995; Crotty). Actors come to understand the world through v repeated transactions with it, therefore, knowledge is necessarily partial and provisional (Atkinson & Housley, 2003). Knowledge as it relates to symbolic interactionism is created through a process of interpretation of the world by which one is confronted. As such, knowledge takes the form of perception and meaning which in turn is used as the basis for directing human action. In describing the nature of human action, Blumer (1969) states: .. .the human individual confronts a world that he must interpret in order to act instead of an environment to which he responds because of his organization. He has to cope with the situations in which he is called on to act, ascertaining the meaning of the actions of others and mapping out his own line of action in the light of such interpretation (p. 15). Taken into account in this process of human interpretation are such practical factors as individual wishes and wants, objectives, available means for achievement of goals, actions and anticipated actions of others, self-image, past experiences, and the anticipated outcome of a given line of action (Blumer). Central to the process of interpretation in which social actors are said to be engaged is one's propensity toward self-knowing, self-consciousness, self-awareness, and self-communication or what Mead (1934) considered 'possessing a self. According to Blumer (1969), through exploration of self a human being becomes an object, like all other objects, of his own action. Blumer writes: ".. .he is an object to himself; and he acts toward himself and guides himself in his actions toward others on the basis of the kind of object he is to himself (Blumer, 1969). Self in this instance is otherwise referred to as the 'knower' or what Mead referred to as the T . The T occupies the position of the subject and is experienced as being in or ready for purposeful 47 action at any given moment (Blumer & Morrione (Ed.), 2004). The essence of the T then is that part of the individual that is impulsive and spontaneous and provides a sense of initiative or propulsion to act (Charon, 1979). Self is also viewed as a separate social object because the individual comes to see the self in interaction with others. In other words, one's self is pointed out and defined socially (Charon, 1979). As object, self can be identified, judged, communicated with, directed, and manipulated. Charon commented that "individuals are able to act toward themselves in the same sense that they are able to act toward other objects pointed out to them in interaction" (p. 70). Self occupies the subjective position of that which is known or what Mead (1934) refers to as 'Me'. The 'Me' experienced as object is viewed from the perspective of the community or generalized other (Blumer & Morrione (Ed.), 2004). The complete self, according to Mead (1934), encompasses both the T and the 'Me'; each responds to the other in terms of viewing one's self in action. The relationship between T and 'me' is described by Mead as: The T is the response of the organism to the attitudes of the others; the 'me' is the organized set of attitudes of others which one himself assumes. The attitudes of the others constitute the organized 'me', and then one reacts toward that as an T (p. 175). The ongoing self-reports that result from the interaction between the T and the 'Me' serve as the basis for self-evaluation, self-definition, and for guiding or controlling one's conduct (Blumer & Morrione (Ed.), 2004; Lai, 1995; Reynolds & Herman-Kinney, 2003). One's definition of self is also affected by imagination and imagination of self is influenced by available imagery (sources of mass media) which ultimately positions the self as an object in opposition to a socially constructed other (Lai, 1995). 48 The view that human beings possess a complete 'self capable of thinking about and communicating or otherwise interacting with themselves, as they do with all other human beings, suggests that the notion of self is a process. Blumer argues: "The fact that individuals may indicate something to themselves and respond to their indication -or in other words communicate with themselves - signifies an ongoing process rather than a contemplative regarding of themselves as an object" (Blumer & Morrione (Ed.), 2004, p. 65). As a mechanism for self-interaction then, the essence of 'self as process is considered to be cognitive in nature -an internalized conversation which constitutes thinking and reflective processes and takes the form of reflective knowing (Atkinson & Housley, 2003; Charon, 1979; Reynolds & Herman-Kinney, 2004). With regard to the second epistemological principle that what we know is determined by what we are able to observe, the relationship between actor/ knower and what is known is based on actors' interpretations of their interaction with the social world as a dynamic process (Crooks, 2001). In symbolic interactionist thought, knowledge is derived from interpretation of both overt and covert action. Overt action is seen as one's manipulation of their environment. Covert action is that which is considered 'mind' activity, and the interplay between the two (Charon, 1979). The notion of overt and covert action and the interplay between the two is best captured by Mead's (1938) four stages of the act. In the stage of impulse, the actor experiences a state of disequilibrium which causes him/her to define self and the situation, which in turn is followed by an impulse or generalized inclination to act. This impulse does not in any way dictate the form or direction of the action, only that action of some sort will occur. Human beings are constantly experiencing various degrees of disequilibrium forcing them to actively respond to or deal with various aspects of their environment (Blumer & Morrione (Ed.), 2004; Mead, 1938). The impulse stage sensitizes individuals to those environmental stimuli that are most relevant to them and prepares them for the furtherance or development of the act (Blumer & Morrione (Ed.), 2004). The second state, perception, involves a more in-depth analysis of the situation. This process includes interpreting and defining environmental stimuli selected on the basis of their relevance to achieve specific goals and identifying lines of action. Determining a course of action also requires that the actor reflect on and consider both past and future actions thus incorporating knowledge in the form of memory and experience as well as foresight (Charon, 1979). The third stage, manipulation, involves active manipulation of the environment. This stage is considered to be the overt phase of action because of the purposeful manner in which the actor responds toward objects in an effort meet identified goal(s) (Charon, 1979). Finally, the stage of consummation (if it occurs) signifies the restoration of tentative equilibrium - tentative in the sense that it is only achieved and/or maintained in the absence of other act sequences that result in a constant stream of action (Charon). As a social process, human action incorporates both covert and overt phases of action construction (Charon, 1979). Covert action relates to stages involving defining the self and the situation and establishing goals and determining lines of action. Overt action involves the manipulation of the environment to achieve identified objectives aimed at restoring equilibrium or, in other words, managing with the situation in which one finds him or herself. As mentioned previously, the symbolic interactionist view of human meaning and action has implications in terms of how human interaction is understood. Similarly, it has implications for how human action is or ought to be studied. 50 3.5 Methodology Methodology represents the principle ways in which researchers act on their environment (Denzin, 1989). Methodology is, however, contextual. Methodology is selected and implemented based on the philosophical stance or theoretical perspective that lies behind the choice (Crotty, 1998). Theory, according to Denzin, provides "order and insight to what is or can be observed" (p. 4). Methodology also provides rationale for the choice of research methods and the form in which the methods are employed in terms of yielding data that represent different features of reality and make the research public and reproducible by others (Crotty; Denzin). As discussed previously, symbolic interactionists maintain that reality exists in the empirical world and is to be discovered in the examination of that world (Blumer, 1969). Knowledge is believed to arise from and be based on principles of exploration and interpretation (Blumer; Atkinson & Housley, 2003). Scientific knowledge is "understood not in terms of abstract philosophical prerequisites, but in terms of human engagement with the world about us" (Atkinson & Housley, p. 122). Symbolic interactionism is a practical approach to scientific study in which scientific methods are considered mere instruments designed to identify and analyze what is actually going on in real life, particularly in terms of reflecting human interpretation of the empirical world (Blumer, 1969). As an empirical science, therefore, it is important that symbolic interactionism respect the empirical world under investigation by fitting its problems, guiding conceptions, procedures of inquiry, techniques of study, and theories to that world (Blumer). Ultimately, the symbolic interactionist perspective guides research activities in such a way that they "yield verifiable knowledge of human group life and human conduct" (Blumer, p. 21) or in the words of Denzin (1992): "make the world of lived experience directly accessible to the reader" (p. XIII). 51 Of central methodological importance in examining the empirical world from a symbolic interactionist perspective is the 'act' or 'action' itself. Since action is comprised of observable properties, empirical research techniques applied to the actual social setting are of some use in capturing qualities of human behavior (Blumer, 1969). As mentioned previously, however, symbolic interactionists maintain that action is also comprised of covert qualities - that is human behavior in relation to self and others occurs in a given context that can neither be observed nor predicted with any degree of certainty. Consequently, observational techniques alone may be limited by the lack of meaning they are able to convey. To grasp the meaning of human action, the symbolic interactionist perspective requires the use of methods that are amenable to observation of overt behavior, as well as those approaches that promote understanding about the thinking processes that direct action. As argued by Charon (1979), the researcher employs methods that promote understanding about the manner in which people come to define actual situations, how they develop and use perspectives, change perspectives, role-take, apply their past, problem solve, converse with self, and decide on a line of action. In short, investigators must seek to examine the everyday world as it is lived and experienced by the actor or actors. The everyday world or 'empirical world', according to Blumer (1969), constitutes the actual group life of human beings as denned by what actors individually and collectively experience and do. Blumer states: "the life of human society or of any segment of it or of any organization in it or of its participants consists of the action and experience of people as they meet the situations that arise in their respective worlds" (p. 35). Hence, the empirical world represents all aspects of human life and society. Although it is important to take into account social structures in terms of how they influence human perceptions and behavior, because symbolic 52' interactionism emphasizes social process as opposed to social structure, study of the empirical world necessarily incorporates the relation between presents, pasts, and futures (Hall, 1987; Lai, 1995). Returning to the goal of understanding human action, researchers must set their aims to understand how humans act in the present by applying past experience and future plans. Charon (1979) described the interrelationship between past, present, and future as being integral to the individual's ability to define the situations they face. The past, described by Mead (1938) as 'experience' in the form of memory, and 'hypothetical' future (one's plan) does not dictate human action, but both are considered in the process of undertaking action. This implies that symbolic interactionism as a scientific perspective must strive to explain the cause of human action - that is, how individuals think and define situations; how present actions are affected by past experiences and future plans; how they define and respond to others; and how they solve the problems or situations which confront them (Charon). Consequently, while observation in real settings is important, incorporating data collection methods that employ direct communication (such as semi-structured interviews) is critical to obtaining the emic perspective (Morse & Field, 1995), or, in the case of a study examining how participants manage relationships, 'getting inside the heads of nurses, patients, and family members. Dialoguing directly with participants will enable me to become more aware of their perceptions, feelings, and attributes, which will enhance my ability to interpret their meanings and intent (Crotty, 1998). Mead captures the importance of dialogue to the research process as follows" "Observation, hypothesis, and experiment lie.. .in the biographies of the individual.. .and so does the emphasis of attention which marks analysis and the process of so-called logical thinking" (p.67). 53 To fully grasp the other's perspective, the investigator is compelled to take, to the best of his/her ability, the standpoint of those being studied or to put oneself in the place of the other (Crotty, 1998). This action requires a commitment to not only actively enter the world of interacting individuals but to do so as freely as possible with minimal preconceptions, assumptions, or a priori hypotheses (Denzin, 1989). By putting one's own perspective on 'hold' the researcher is better positioned to capture the voices, emotions, and actions associated with the lived experience of others and to strive to understand the meaning that people assign to the situations they encounter and the ways in which meaning ultimately guides action (Crotty; Denzin, 1989; Denzin, 1992). Finally, given that symbolic interactionism emphasizes social human action as processual in nature, the researcher must employ strategies that promote understanding of the dynamic and contextual characteristics of social organization and its influence on human interaction. This means not only taking into account the many layers of contextual and perceptual aspects of reality but also acknowledging its dynamic quality (Mead, 1938). Methodologically then, incorporating strategies that take into account the contextual and systemic characteristics of the environment, such as participant observation in various acute care settings, would be important to the process of understanding how such factors affect nurses' practice and patients' and families' efforts to manage relationships. Moreover, implementing strategies aimed at seeking to understand how participant behavior fits or does not fit with their articulation of meaning associated with contextual and systemic characteristics within the environment is also critical. The dynamic nature of human social action also speaks to the symbolic interactionist view that knowledge, as in the case of scientific findings, is also dynamic. Mead (1938) captures this idea when he states: 54 The fundamental assumptions involved in the account of knowledge that I am presenting are: that it is a process of finding something that is to take its place in a world that is there, which world that is there is the presupposition of the undertaking that we call "knowledge": that the world that is there is a temporal world i.e., that it is continually passing,.. .that the world is therefore continually ceasing to be as it passes into the world of the following moment.. .(p. 64). The goal of symbolic interactionist research, therefore, is the development of modifiable theories about human social interaction patterns that rest on the behaviors, definitions, meanings, and attitudes of those studied (Denzin, 1989). In explaining how nurses working in acute care hospital settings and patients and family members manage relationships when providing care, it would be important for me to account for variables, such as individual meaning and perception, personal factors, and contextual and systemic conditions that affect interactive patterns over time. 3.6 Summary As a theoretical perspective on life, society, and the world, symbolic interactionism directs interpretive research in a way that focuses on the meaning of basic social interaction from the perspective of the actors. Strongly influenced by pragmatist philosophy, which maintains that people are emergent beings engaged in ongoing, active interaction with their environment, symbolic interactionism rests on three main premises related to human action. The premises are as follows: (1) people individually and collectively act based on the meaning they assign to symbols or objects in their life; (2) the meaning of symbols is derived from or arises out of social interaction; (3) meanings are perceived and modified through a process of interpretation. Symbolic interactionism emphasizes the importance of exploration and interpretation of human engagement in the empirical world. The empirical world is considered as all aspects and contextual layers of what actors individually and collectively experience and do as they confront the situations that arise in their respective worlds. Symbolic interactionism, therefore, supports 5 5 exploration and interpretive analysis of experiences, actions, and variations across time and context in natural settings. I am directed to employ methods that promote understanding about how people define actual situations, assign meaning and develop and use perspectives, role-take, problem solve, converse with self, decide on a line of action, and act. Given that my primary research goal, as guided by symbolic interactionism, is to obtain and describe others' perspectives and behaviors with regard to social interactive processes, I will use methods such as observation, reflection, interaction and dialogue in naturalistic settings. Thus, the strategy of inquiry employed in this study is the grounded theory method which will be described in-depth in chapter four which follows. 56 CHAPTER FOUR 4 Strategy of Inquiry: Grounded Theory The purpose of this chapter is to describe the strategy of inquiry employed in this study. The strategy of inquiry connects the theoretical perspective of a study to a particular research design which in turn guides the choice and use of specific methods for collecting and analyzing data to achieve a desired outcome (Crotty, 1998; Denzin & Lincoln, 2000). The primary goal of grounded theory guided by symbolic interactionism is to explain social interactive processes. A detailed account of the grounded theory method, sampling, ethical considerations, recruitment, data collection, data analysis, and rigor will be described in this chapter. 4.1 Grounded Theory as Informed by Symbolic Interactionism Symbolic interactionism is widely cited as the theoretical foundation or perspective upon which grounded theory is based (Benoliel, 1996; Hutchison & Wilson, 1993; Morse & Field, 1995). As previously discussed, symbolic interactionism is concerned with understanding people's behavior from their perspectives. Their behavior is influenced by social interactions, interaction with self, and the socio-cultural environment or context in which they live. Grounded theory as informed by symbolic interactionism, therefore, aims to represent the complex interactional processes as theoretical explanations that are inherently relevant to the practice world from which they emerged (Hutchison & Wilson). In the tradition of interpretivist philosophy, grounded theory researchers who use the symbolic interactionist perspective regard knowledge as influenced by temporal, socio-cultural, and subjective conditions, which inform multiple symbolizations of reality. Consequently, the social and interpersonal context of the actors, as well as their intentionality and conscious construction of meaning as expressed through symbols and social interactions, is of central 57 concern (Benoliel, 1996). Based on the importance of seeking meaning and understanding from the actor's point of view, it is critical that the grounded theorist enter the field with as few pre-determined ideas or prior hypotheses as possible. The researcher's mandate is to remain open to what is actually happening as opposed to filtering or framing data (at all stages of collection and analysis) through his or her biases (Glaser, 1978). As stated by Glaser "The goal of grounded theory is to generate theory that accounts for a pattern of behavior which is relevant and problematic for those involved" (p. 92). Central to the grounded theory method informed by symbolic interactionism is the assumption that interpretations of reality are socially constructed. Two key implications of this assumption are that a) people sharing common circumstances likely share a specific psychosocial problem that may not necessarily be articulated and, b) in response to this common problem people order and make sense of their environments (Hutchison & Wilson, 1993). The aim of grounded theory research, therefore, is to develop core variables that explain patterns of behavioral responses to a main concern or problem actors try to manage in particular social settings. For this study, the goal was to understand and explain individual and collective perceptions of and patterns of actions/behaviors of nurses, patients, and family members as they manage relationships during care giving within a specific context - the acute care hospital setting. Since little is understood about nurse-patient-family relationships as they develop and evolve in contemporary acute care settings, grounded theory was an appropriate means of inquiry. When little is known or understood about a topic, grounded theory is particularly useful because it focuses on analysis of everyday meanings and patterns of behavior to generate theoretical explanations (Dempsey & Dempsey, 2000; Glaser, 1978; Hutchison & Wilson, 1993). Capturing 58 meaning in order to explain behavioral patterns of a specific group directs the investigator toward methods of seeking actors' interpretations or meaning-making in terms of personal, relational, and contextual features - a hallmark of symbolic interactionism. 4.2 Grounded Theory - An Overview Grounded theory is a research method that is used to systematically generate theory from data (Glaser & Strauss, 1967; Glaser, 1978). The study design is intended to identify social problems and the social-psychological and social-structural processes that arise as humans adapt to situations in which they find themselves (Benoliel, 1996; Crooks, 2001). Morse and Field (1995) asserted that the primary goal of grounded theory is to explain human behavior. Barney Glaser and Anselm Strauss described grounded theory in 1967. According to these authors, substantive theories are considered 'grounded' because they are systematically developed through a process of data collection from multiple sources, theoretical sampling to develop and refine emerging categories, and constant comparative analysis that occurs simultaneously with data collection. As a result, hypotheses and concepts that make up the theory are constructed from the data (Glaser & Strauss, 1967). Data are collected in natural settings using techniques such as semi-structured interviews and participant observation. Analysis focuses on meanings, experiences, and context. The analysis generally produces a core variable as well as codes, concepts, categories, and hypothesized relationships among categories. The core variable often represents a social process that offers a tentative explanation of the phenomenon being studied (Hutchison & Wilson 1993; Jacelon & O'Dell, 2005). In short, "the process [of grounded theory], generates theory that fits the real world" (Glaser, 1978, p. 143). 59 Since its inception, grounded theory has developed in diverse ways (Benoliel, 1996; Boychuck & Morgan, 2004; Denzin & Lincoln, 2000; Heath & Cowley, 2004). Most notably, the difference in opinion that evolved between Glaser and Strauss resulted in what is now commonly regarded as two different typologies of grounded theory methods: the Glaserian model and the Straussian model (Stern, 1994). While it is not within the scope of this chapter to recount the details of what Melia (1996) refers to as "a head on clash between Glaser and Strauss", it is important that when using grounded theory I declare my interpretation of grounded theory which I chose to guide my research, which is the 'Glaserian' model. The divergence of opinion between Glaser and Strauss and Corbin became public with the release of Glaser's book Emergence vs. forcing in basics of grounded theory analysis (1992). In this text, Glaser expressed his objections to Strauss and Corbin's (1990) representation of grounded theory; his objections spawned the ongoing debate over which method captures the essence of grounded theory (Hutchison & Wilson, 1993). While Strauss and Corbin argued that they have refined the grounded theory method by providing a more straightforward approach to data analysis (Hutchison &Wilson), Glaser (1992) maintained that Strauss and Corbin's version is not grounded theory but full forced conceptual description. At the heart of the debate lies procedural differences between the Straussian and the Glaserian model that center on differences in data analysis (Heath & Cowley, 2004; Melia, 1996). What follows is a brief overview of three main methodological points of departure between Glaser and Strauss and Corbin. The first point relates to the nature of the research question. As described by Melia (1996) and Heath and Cowley (2004), Strauss and Corbin considered the research question to be a statement that identifies the phenomenon to be studied and what is known about the subject. Glaser (1978; 1992) on the other hand, argued that the research question should emerge and evolve as data are collected and analyzed rather than being predetermined. The notion of emergence follows from Glaser's commitment to the importance of theoretical sensitivity, or in other words, the analysts' ability to 'remain open to what is actually happening' (1978, p 3). While Glaser and Strauss both agreed that prior understanding of the general problem area contributes to one's ability to be open to a wide range of possibilities, Strauss maintained that both theoretical sensitivity and the process of generating hypotheses can be stimulated through initial and purposeful use of both the self and the literature, without introducing researcher bias (Heath & Cowley, 2004). Contrary to Strauss' position, Glaser (1978) stressed that, in order to remain open to the data, it is important to enter the research setting with as few "logically deducted, a prior hypotheses as possible" (p.3). Glaser argued that literature should be incorporated later in the process once the theory is considered sufficiently grounded and developed. At that point the literature is considered to be ".. .part of the data and memos to be further compared to the emerging theory to generate even more dense [sic], integrated theory of greater scope" (Glaser, p. 7). Glaser further argued that, in order to remain open to the emerging theory, the researcher ideally begins with a sociological perspective about a general problem area. I agree with Glaser that such a perspective could take the form of "some combination of a clear question or problem area in mind, a general perspective, and a supply of beginning concepts and field research strategies" - an approach that, while somewhat limiting, is still receptive to the emergent (p. 45). The second point generally relates to the process of data analysis, particularly as it pertains to initial coding. Strauss and Corbin advocated a more mechanical, highly structured procedure of breaking down and conceptualizing the data, for example, axial coding operations (Melia, 1996). Glaser promoted theoretical coding techniques that involve analytic processes such as constant 61 comparison of incidents and concepts within the data for the purpose of identifying similar patterns of incidents that can be categorized (core categories and basic social processes) until saturation occurs (Boychuk Duscher & Morgan, 2004; Glaser, 1978; Heath & Cowley, 2004). There is also contention about the end product, that is, whether theory is forced or generated (Heath & Cowley). Boychuk Duscher and Morgan captured this point when they wrote: " Glaser's contention that Strauss and Corbin's questions are highly structured, reductionistic, and intrusively deconstructive in nature is fundamental to his allegation that they force a full conceptual description of the data rather than articulate emergent theory" (p. 608). I support Glaser's criticism that claims the Straussian approach forces data to fit preconceived conceptual categories as opposed to letting the data 'speak for itself (Glaser, 1978; Glaser, 1992). 4.3 Research Design 4.3.1 Purpose. As previously stated, I used the 'Glaserian' model for my study. By the 'Glaserian' model, I refer to the model of grounded theory first described by Glaser and Strauss (1962) and later elaborated by Glaser (1978) - informed by the theoretical perspective of Symbolic Interactionism. The ultimate goal of this study was to construct a core category that explains the general phenomenon under investigation - namely, the nature of managing relationships between nurses, patients, and family members in the acute care hospital setting. A core category represents the main theme associated with the primary problem of people in the setting under investigation and explains, by summing up in a pattern of behavior, what is going on in the data (Glaser, 1978). In order for a theme to be classified as a core category it must meet the following criteria: it recurs frequently; it maintains a position of centrality by explaining much of the 62 variation in the data; it relates meaningfully and easily to other categories; it takes longer to saturate, and; it has clear implications for formal theory position (Glaser, 1978). A core category may take the form of a basic social process (BSP) - the main difference between a core category and a BSP being that a BSP is "processural", meaning that is has two or more clear emergent stages (Glaser, 1978). According to Glaser "BSP's...are fundamental, patterned processes in the organization of social behaviors which occur over time and go on irrespective of the conditional variation of place" (p. 100). As Benoliel (1996) explained, interpretation of a basic social process results in an explanation of the phenomenon being studied and provides the core of a substantive theory. Basic social processes are further categorized as basic social psychological processes (BSPP) and basic social structural processes (BSSP), although the two forms of categories are interrelated. Basic social structural processes represent a set of structural conditions that constrain, facilitate, or simply create the social structure under which social psychological processes occur (Glaser). Generally, basic social psychological processes (BSPPs) are implied by basic social processes (BSPs); when generating a social structural process, however, the researcher must clearly state it as such by using BSSP (Glaser). Depending on the nature of the emerging theory, the basic social process may emphasize the BSPP, the BSSP, or both. 4.3.2 Questions. Three research questions, which reflected a broad perspective about acute care nurse-patient-family relationships, were developed to guide this study. Given that the main purpose of the study was to explain how nurses, patients, and family members managed relationships when providing care in acute care hospital settings, these research questions focused on structural, 63 processual, and interactional features of nurse-patient-family interactions from the perspective of participants. The specific research questions were stated as: 1. How do nurses, patients, and family members perceive their management of relationships in order to plan and provide patient care in the acute care hospital setting? 2. How do contextual and systemic features or structural conditions affect nurses', patients', and family members' efforts to manage relationships to plan and provide patient care? 4. How do personal factors affect nurses', patients', and family members' perceptions of their efforts to manage relationships to plan and provide patient care? The questions were based on my sensitivity to terms such as collaboration and decision-making which have been associated with nurse-patient-family relationships. These sensitizing concepts proved to be irrelevant based on the 'local' concepts (those used by participants) that began to emerge. I modified my questions during the study to stay open to the data (Glaser, 1978). 4.3.3 Initial purposive sampling. A key feature of the grounded theory method for data collection and interpretation is the recursive, process-oriented analytic procedure characterized by constant comparison and theoretical sampling (Greene, 1998; Locke, 1996). Constant comparison involves the process of comparing and contrasting "every piece of relevant data with every other piece of relevant data" for the purpose of identifying patterns or instances of phenomenon that seem similar or dissimilar (Morse & Field, 1995, p. 158). Theoretical sampling, which is based on collecting, coding, and analyzing data, refers to making decisions about which data should be collected next to develop the theory as it emerges (Dey, 1999; Glaser 1978; Kendall, 1999). Since the emerging theory guides the participant selection process, it is impossible for the researcher who uses theoretical sampling to know in advance precisely what to sample for and 64 where it will lead torn/her (Glaser, 1978); however, the researcher has to start somewhere. Consequently, Glaser maintained that initial decisions must be made on the basis of a general sociological perspective about the substantive area as opposed to a preconceived problem or hypothesis. Such a process requires the investigator to begin with purposive sampling. Purposive sampling, which occurred at the initiation of my grounded theory study, was based on a general sociologic perspective about a substantive area which takes the form of a somewhat limited framework of related or "local" concepts (Glaser, 1978; Glaser & Strauss, 1967). At this stage, I purposefully selected participants from the field of inquiry based on their willingness and ability to generally speak to the phenomenon of interest as specified by Glaser. My initial purposive sampling decisions influenced the selection of participants who were involved in particular incidents, events, and activities in particular settings. As I collected, coded, and analyzed the data, I shifted the conceptual framework to reflect what was happening in the data and then selected participants based on the contribution they were likely to make to the emerging theory. According to Morse (1999) "it is this process of selecting that ensures that the theory is comprehensive, complete, saturated, and accounts for negative cases" (p. 5). J continued purposive sampling and constant comparison of data until beginning categories began to emerge. Since the goal of my research study was to explore how nurses, acute medical/surgical patients, and their family members managed their relationships with each other during the provision of care, I designed my sampling to include each of these groups (nurses, patients, and family members) from medical, surgical, transitory care, and intensive care units. Although I initially planned that recruitment would occur from nursing units within three cornmunity hospitals in the Fraser Health Authority of British Columbia, I later obtained permission to expand recruitment to include two additional hospitals within the region. Due to saturation and constraints associated with my dissertation, I only sampled from four of the five hospitals. Although I considered obtaining data from nurses, patients, and family members interacting simultaneously around patient care as ideal, I anticipated and ultimately confirmed that the unpredictability of family visits and the complexity of acute care settings would make obtaining data from interacting groups of nurses, patients, and family members very difficult. Consequently, I collected data from nurse, patient, and/or family member participants who were independent of each other. In other words, I asked nurses to speak about their general experience and perceptions about managing relationships between themselves and patients and family members when planning and providing care. Similarly, I interviewed patients and family members about their general experiences and perceptions while managing relationships with nurses. When I asked participants to reflect on their experiences and perceptions, I asked them to refrain from revealing information that would disclose the identity of the nurse(s), patient(s), and/or family member(s) discussed. I based inclusion criteria for participation on the nature of the particular acute illness experienced by the patient by considering its documented prevalence within acute care hospitals in Canada, average length of hospital stay, and life threatening/life altering effect of the illness for patients and family members. 4.3.4 Inclusion/exclusion criteria for patient participants. I identified initial purposive sampling criteria as adult patients (defined as being 19 years and over), admitted to one of four community hospitals within the Fraser Health Authority of British Columbia, Canada, with a diagnosed acute illness episode. I based determination of an acute illness episode on the Canadian Institute of Health Information (2002) report of diagnoses most likely to result in acute care hospitalization across Canada in 1999/2000 and the British 66 Columbia Ministry of Health Planning (2002) report on the increased incidence of respiratory related diseases, unintentional injuries, colorectal cancers, and diabetes. Thus, I considered diagnoses to be acute illness episodes when they were: cardiac disease such as heart attack or cardiovascular accident (stroke); respiratory disease including lung cancer; digestive disease including stomach, gastrointestinal or colorectal cancers; other cancers considered to be in an acute stage of initial diagnosis or recurrence e.g. leukemia; unintentional injury such as falls or motor vehicle accidents; and/or acute complications associated with metabolic or other diseases, such as diabetes. While the disease-specific length of stay statistics indicated that the average minimum length of stay for an acute illness ranges from 5.5 days for respiratory diseases to 9.5 days for malignant and benign cancers (Canadian Institute of Health Information), the actual length of stay for acute illness may be substantially less. Since I was concerned with understanding and describing how nurses, patients, and family members manage relationships, which occurred over time, I excluded patients and family members of patients hospitalized for less than twenty four (24) hours from the sample. I also excluded patients who were unconscious, semi-conscious, or otherwise unable to communicate verbally for reasons associated with disability, medication, treatment, and language barriers. If I interviewed a patient who was independent of any particular patient-nurse-family member triad or dyad, I asked him or her to reflect on his or her interactions with nurses, and between family members and nurses in which he or she participated. 4.3.5 Inclusion/exclusion criteria for nurse participants. I included nurse participants who were registered with the College of Nurses of British Columbia and employed by one of the community-based hospitals located within the Fraser Health Authority of British Columbia, Canada. These nurses were employed on a full or part-67 time basis on the participating medical, surgical, transitory, activation care, or intensive care units. In order for a nurse to meet my qualifications for being interviewed about managing relationships in the nurse, patient, and family triad, he or she must have been assigned to that patient for a minimum of (not necessarily consecutive) sixteen hours or the equivalent of two-eight hour or 1.3 twelve-hour shifts. If I interviewed a nurse participant who was independent of any particular patient-family triad or dyad, I asked him or her to reflect on interactions, in general, with patients and family members, to whom she or he had been assigned for at least the period of time stipulated. Registered nurses who were employed by the hospital on a contract or casual basis were not eligible to participate in the study. I also excluded registered psychiatric nurses, licensed practical nurses and other 'nurse' designations, such as nurse managers. 4.3.6 Inclusion/exclusion criteria for family member participants. I included family participants who were 19 years or older and identified by the patient, the nurse caring for the patient, or self-identified as a primary source of support for patient care planning and/or delivery. In cases where more than one family member was identified, I included multiple family members in interviewing. If I interviewed a family member who was independent of any particular patient-nurse triad or dyad, I asked him or her to reflect on his or her interactions with nurses and family members who had been patients. I excluded family members who were unable to communicate verbally for reasons associated with disability and/or language barriers. 4.4 Ethical Considerations Initiating and maintaining trust with participating agencies and research participants is the primary focus of any research study involving human beings (Morse & Field, 1995). Trust as a moral imperative requires that the researcher attend to the three ethical principles associated with 68 informed consent: respect for human dignity, beneficence, and justice (Fontana & Frey, 2000; Morse & Field). The principle of respect for human dignity addresses a person's right to self-determination and full disclosure. In the context of research involving human subjects, this means that participants have the right to be fully informed about the nature and consequences of the experiments or studies in which they are involved (Christians, 2000). Both the agency and the subjects must be provided with an understandable and sufficiently detailed verbal and written explanation, including risks and benefits of the study and how they will be involved. The notion of self-determination means that subjects must agree to participate on a completely voluntary basis, free from physical or psychological coercion (Dempsey & Dempsey, 2000), and they have the right to withdraw their participation from the study at any time without penalty or consequence (Morse & Field, 1995). The principle of beneficence relates to the right to freedom from harm and exploitation (Dempsey & Dempsey, 2000). The researcher must treat participants in a way that conveys respect and make every effort to protect their well being. Christians (2000) extends the notion of exploitation to mean opposition to deception that is considered deliberate misrepresentation. In other words, the researcher must not, under any but the most extreme circumstances deliberately misrepresent any aspect of the study. The principle of justice involves the right to be treated in a fair and equitable manner, as well as the right to privacy (Dempsey & Dempsey, 2000). Confidentiality is of extreme importance and must be ensured to protect the participant and the agency from unwanted exposure. This means that the investigator must ensure that the identities of participants will not be disclosed other than in the confines of the consent agreement. The principle of justice also requires that the 69 benefits of the research must be accessible and applied equally to all members of society (Dempsey & Dempsey; Liaschenko, 1999). Because protecting the rights of participants is of critical importance, I implemented initiatives to address the principles of human dignity, beneficence, and justice in this study. I began with securing ethical approval and any extensions from various governing bodies associated with the study. I then negotiated access and obtained informed consent. I now present an overview of these measures and associated issues. 4.4.1 Obtaining ethical approval. In compliance with the Faculty of Graduate Studies at the University of British Columbia (UBC), I sought ethical approval from the University of British Columbia Behavioural Research Ethics Board (Appendix A - Certificate of Approval), the Fraser Heath Authority (Simon Fraser Area) Clinical Investigation Committee (Appendix B), and the South Fraser Health Authority Regional Research Review Committee (Appendix C), prior to commencing my study. During the course of data collection, the two regional health authorities merged to form the Fraser Health Authority. Obtaining ethics approval from the appropriate governing bodies is a necessary measure to protect the rights of participants as it requires that the research protocol, including documents such as the information letters used to describe the study and participant rights (Appendix E), consent forms (Appendix F), and demographic questionnaires for nurses (Appendix G), family members (Appendix H), and patients (Appendix I) be reviewed and accepted. I obtained ethical approval from the University of British Columbia Behavioural Research Ethics Board for a one year term commencing November 2003. The South Fraser Health Authority Regional Research Committee and the Fraser Health Authority (Simon Fraser Area) Clinical Investigation Committee granted approval in October 2003. 70 4.4.2 Ethical approval extension. I encountered several common delays associated with the process of negotiating access and commencing recruitment, including difficulty making contact with administrators due to vacation and work demands. The most significant barriers included the closure of a community hospital in the early spring of 2004 followed by the Health Employees Union (HEU) job action that began in the late spring of 2004. Both events resulted in considerable disruption to nursing units and the hospitals where I planned recruitment, which meant further delays in negotiating entry to the field. Consequently, the recruitment period extended beyond the approved twelve month period and an application for extension became necessary. Since data collection had been completed at the South Fraser Health Authority site (N= 4 sites), I submitted and received a request for extension from the University of British Columbia Behavioural Ethics Review Board (Appendix A) and the Fraser Health Authority Clinical Investigation Committee (Appendix B). 4.5 Procedures 4.5.1 Negotiating access. I negotiated and renegotiated access to the field. Negotiation and renegotiation includes such issues as where, when, and with whom the research study will take place (Morse & Field, 1995). To negotiate entry to each site I telephoned and/or made e-mail contact with the hospital administrator and submitted ethical approval certificates and an information letter and consent form. I obtained permission from administrators to contact nursing unit managers. I contacted managers via telephone and/or e-mail and provided information letters and consent forms. I met with unit managers to discuss my specific unit access and recruitment needs. The unit managers identified units under their supervision that provided care for the patient categories outlined in my inclusion criteria with regard to admitting diagnoses. After identifying 71 appropriate units, the managers informed the patient care coordinators about the study, requested that they cooperate with me, and provided me with patient care coordinator contact information. I contacted patient care coordinators to discuss details of the study, such as specific recruiting strategies, and to enlist the assistance of their staff nurses, nurse educators, and unit clerks. 4.5.2 Participant recruitment strategies. Once I obtained access, I began recruiting nurses, patients, and family members. In keeping with the right to self-determination - that is the right of prospective participants to decide voluntarily whether to participate in the study without risk of penalty or prejudicial treatment (Loiselle & Profetto-McGrath, 2004), I avoided any form of coercion when distributing information about the study and an invitation to participate. Given the possibility that I could be perceived as having authority over potential participants, I requested staff nurses or patient care coordinators on the unit to distribute information letters (Appendix E) and/or otherwise inform peers, patients, and family members about the study. Those employees ensured that potential participants received information letters so that they could read them at their convenience. The information letter provided details about the study including: information about the principal and co-investigator, the purpose of the study, study procedures, participants' rights, and contact information should participants have any concerns about the study. Upon receipt of the information letter, participants initiated contact with me. Individuals who decided they were interested in participating or required further information or clarification contacted me to set up a meeting. I found that recruitment was slow when patient care coordinators and staff nurses forgot to distribute the information letters to patients and family members. On some occasions when potential participants received the letter and indicated interest, staff members did not follow up. 72 After several follow-up calls to the units failed to yield any participants, I made adjustments to the recruitment approach. Upon consultation with the patient care coordinators I decided that information letters would be distributed to patients (who met the criteria), family members, and nurses during times when I was on the unit. This arrangement allowed me to be immediately available to meet with potential participants who expressed an interest in being interviewed or wanted more information. Decreasing time between receiving the letter and initiating direct contact with me accelerated the recruitment process and increased my chances to secure participants. I planned my availability in order to avoid 'busier' times on the unit, such as early mornings, and late afternoon when patients were typically napping. Nurses tended to have more free time in the early afternoon, family members were more likely on the unit, and patients were more likely awake and available to talk. In the case of nurses who wished to participate, the hospital, through the efforts of the unit manager and/or patient care coordinator, provided coverage so that the participant nurse could be freed from her (all nurse participants in this study were female) duties - in some cases for up to forty-five minutes. To confirm participation in my study, I obtained informed (written and verbal) signed consent (Appendix F). Because I provided a detailed information letter and a consent form that reiterated the nature of the study and rights of participation and reviewed these documents verbally with individual(s), I maintained the requirement of full disclosure. Participants who gave informed consent retained the right to withdraw from the study at any time without question. Of the forty participants (N = 40) recruited, none withdrew from the study and I received no complaints about the study or study procedures. 73 I maintained confidentiality by omitting participant names or any other identifying information (e.g. diagnosis, hospital) on audio-tapes or any study documents other than the consent forms which required a signature. I identified audio-tapes and documents by using a letter/number combination (2 digit code) and kept all documents and tapes in a locked filing cabinet. I protected computer data with a password. During data collection, I respected confidentiality by conducting interviews in areas that were quiet, free from distraction, and as private as possible. Most interviews were conducted in an interview/conference room adjacent to the nursing unit(s). If such a room was not available or patient participants could not leave their beds, I conducted interviews in patient rooms. In most cases, the patient and/or patient and family member were the only ones in the room at the time of the interview. In the small number of cases where the room was shared with others, the other patient was out of the room or appeared to be sleeping. I drew the patient participant's curtain to provide privacy and spoke softly to prevent the conversation from being overheard. 4.5.3 Data collection. I obtained data for this study from a total of forty participant interviews with patients (N = 13), nurses (N = 17) and family members (N = 10). While the majority of participants were interviewed individually, nurse participants N-04 and N-05 were interviewed together, as were family members F-05 and F-06. As well, on three occasions a patient and family member participant were interviewed together. I not only transcribed each interview, but also captured my thoughts, impressions, and observations about them in field note data. In addition, I undertook a total of thirty-three hours of participant observation and wrote field notes describing those experiences. 74 Theoretical sampling guided my data collection process. According to Glaser (1978), theoretical sampling is itself a process "whereby the analyst jointly collects, codes, and analyzes his [her] data and decides what data to collect next and where to find them, in order to develop his [her] theory" (p. 36). Critical to this process is that data are collected on the basis of a general sociological perspective about an area of interest or concern about a population (Glaser; Glaser & Strauss, 1967). While predetermination of the direction that data collection will take in grounded theory research is not possible or even desirable, initial data collection strategies were aimed at identifying a partial framework of 'local' concepts about the social organization and action under consideration, as suggested by Glaser, (1978). Theoretical sampling requires a multi-faceted investigation in which there are "no limits to the techniques of data collection the way they are used, or the types of data acquired" (Glaser & Strauss, 1967, p. 65). Glaser and Strauss argued that flexibility in the form of data and methods used to collect them is especially important when the researcher must work around the structural conditions (schedules, restricted areas, and varying individual perspectives) of groups. Also, they argued that the more 'slices of data' obtained, that is, the more data collection techniques used, the richer the information will be and the more depth the emerging categories will have (Glaser & Strauss). Data collection is primarily intended "to gather extensive amounts of rich data with thick description" (Charmaz, 2000, p.514). Although I anticipated that the focus of data collection and the strategies to obtain data would evolve on the basis of the emerging theory, I used general techniques for data collection which included: a demographic data questionnaire, semi-structured audio-taped interviews; participant observation, and field notes. I used journaling to capture my personal reflections throughout the 75 process. I will provide a brief description, including the implications of each form of data collection. 4.5.4 Descriptive data. I developed questions to capture participant demographic information such as gender, age, marital status, level of education, and cultural identity. These data provided a description of the sample and evidence that participants met the inclusion criteria (e.g. professional designation, age, reason for hospitalization, length of stay). These data permitted me to discuss the representativeness and transferability of findings to similar populations (Dempsey & Dempsey, 2000). I summarized the data with frequency distributions or means and standard deviations. 4.5.5 Semi-structured interviews. A semi-structured interview strategy is utilized when the researcher has identified main questions but the answers are unpredictable (Morse & Field, 1995). When using the semi-structured interviewing, researchers prepare a conversational guide (Rubin & Rubin, 1995, p. 161) that helps keep the participant focused on the main topic or theme of the interview. Three guides were developed for this study: one for nurse participants (Appendix J), one for patient participants (Appendix K), and one for family member participants (Appendix L). Developing an interview guide required me to think carefully about the main theme and develop questions that linked aspects of the topic together in a logical way that made sense to interviewees. I organized questions for the initial interview guide by themes such as: a) introduction of the topic; b) how the process of managing relationships occurs; c) how contextual and systemic features or structural conditions within the acute care hospital environment facilitate or create boundaries to managing relationships and; d) closure of the interview. I also structured questions so that only one idea was introduced at a time in order to avoid confusion on ( 76 the part of the interviewee (Morse & Field, 1995). Because participants need to understand the intention of the interview questions, I avoided medical terminology. I developed each guide to include general questions relevant to main concepts associated with the substantive area of inquiry (e.g. decision-making, collaboration) to capture various aspects of the process under consideration, namely, how nurses, patients, and family members manage relationships during the provision of care in acute care hospital settings. For example, to nurse participants I asked: Today we are talking about how nurses, patients, and family member(s) collaborate to plan and provide patient care in the acute care hospital setting. Perhaps you could start by describing a recent situation in which you collaborated with a patient and their family member(s) for the purpose of planning and providing patient care? I used open-ended interview questions to encourage participants to express their feelings, opinions, and experiences. For example, in the initial interview guide for nurse participants under the general topic of 'introduction of topic and understanding significance and meaning from the nurses' perspective', I asked: How did the collaborative process related to planning and providing patient care affect you as a nurse? I intended to determine if and how the concept of collaboration fit with the nurses' perceptions of their interactions with patients and/or family members. When participants' responses lacked detail, depth or clarity, I used general probes as prompts to encourage them to elaborate, as suggested by Rubin and Rubin (1995). For example, the guide included brief notes to remind me to probe specifically around the meaning and significance of interactive processes - collaborative or otherwise. Although I prepared interview guides, I did not use them rigidly. I framed the interview using a particular topic but kept boundaries flexible so I could remain open to participants' comments within the limits of the main theme. My questions evolved based on participants' responses. In 77 the example that follows, I was trying to discover a nurse's perspective about the importance of interactions between nurses and family members. I asked: R: When nurses collaborate with family members, how do you think they (the family members) are affected? N: Well, I think they feel part of the care; that they feel part of the healing process with the patient.. .which is good, I mean especially if they are going to be a care giver or help this patient when they go home, they need to know what they can and cannot do. R: How do you think families are affected by the hospitalization of an ill family member? N: I think it is very stressful for them. I think any patient coming into hospital give up so many rights. I mean, we take everything away; as much as you try not to but decisions are made; we're giving you pills. I mean everything that you could do at home has all of a sudden been taken away from you so I think it [collaborating] gives something back to the family -some feeling of control. (N-01, 226-254). At this point I followed the nurse's lead and began the process of trying to tease out the concepts of nurses' power and control over patients and family members and how that affected managing relationships. Power was a notion that I had not previously anticipated. My ability to remain open to the data, in this case the need to follow up on concepts of power and control, was critical to ensuring the relevance of the data to emerging theory as opposed to forcing it into what Glaser and Strauss (1967) referred to as "irrelevant directions and harmful pitfalls" (p. 48). How the researcher conducts the interview is also an important consideration that can contribute to its success in terms of the detail and depth of participants' descriptions (Morse & Field, 1995). I used introductions, thanked participants for their involvement, focused on participants during interviews, asked one question at a time, refrained from interrupting, and conducted interviews at a pace that seemed comfortable to participants. This demonstrated respect as indicated by Morse and Field. Respecting and accommodating moments of silence were also important. Moments of silence may be indicative of new realizations, insights, or may signal that the participant simply requires a few moments to pull their thoughts together (Morse 78 & Field). Richer descriptions result if participants are allowed to continue when they are ready as opposed to being rushed or cut off by a new question (Kvale, 1996; Morse & Field). 4.5.6 Audio-taping. Because accurately capturing the conversation between the researcher and the participant is critical in grounded theory studies, I audio-taped and transcribed the interviews verbatim as soon after the interview as possible. Audio-taping is important because relying on memory is difficult (Silverman, 2000) and it facilitates listening and focusing carefully on questions asked and participants' responses (Morse & Field, 1995). It also "preserves all of the words spoken in the sequence in which they were spoken" (Sandelowski, 1995, p. 373). Transcribing audio-tapes to text provided me with the opportunity to become intimately familiar with the data as indicated by Morse and Field. It also transformed the data into a format that was readily available for further analysis (Sandelowski). Transcribing my own interviews facilitated my abilities to critically examine and evaluate my interview techniques and adjust my approach as required. Both ethical and practical implications must be considered when a researcher plans to audiotape an interview. In compliance with ethical requirements, I informed participants verbally and in writing that the interview would be audio-taped for analysis purposes. I also informed participants that they may request that the tape be turned off at any time. Conveying this information to each participant was necessary to ensure that informed consent was obtained. Pragmatic concerns related to taping included ensuring that the equipment was functioning prior to the interview, carrying fresh batteries and extra tapes, operating the equipment properly, placing the recorder in a strategic position to facilitate the quality of the recording, and most importantly, remembering to turn the recorder on (Baxter & Chua, 1998; Easton, McCorish & Greenburg, 2000; Rubin & Rubin, 1995). Throughout data collection, it was often a challenge for me to find an appropriate surface on which to place the recorder to facilitate the quality of the recording, especially in patient rooms. On one occasion the tape speed was set incorrectly resulting in the need for a second tape. Fortunately, the quality of the recording was not compromised. On another occasion the recorder was set on voice activation mode. This was problematic in that the recorder failed to activate quickly enough to voices so that parts of sentences were not recorded. I transcribed the audio-taped interview immediately to take advantage of my recall to fill in the missing sections of the conversation. 4.5.7 Participant observation. Interpretive research methods that are guided by symbolic interactionism also focus on the interaction between the actors and other actors. According to Mead (1962), understanding of human beings is inferred from what they do, as well as what they think. To fully comprehend the human situation, therefore, researchers need to observe, partake, and interview (Charon, 1995). The primary purpose of participant observation is to get close to the activities and everyday experiences of people in their natural environments (Emerson, Fretz, & Shaw, 1995). Participant observation fits with the grounded theory method guided by Symbolic Interactionism; however, observation as a data collection method is somewhat controversial in terms of the role of the researcher. The primary concern associated with participant observation is the degree to which the behavior or actions of the participants are modified secondary to their awareness of being observed (Mays & Pope, 1995; Talbot, 1995). This is known as the Hawthorne Effect. Emerson, Fretz, and Shaw referred to this phenomenon as "consequential presence" meaning that the researcher's presence may cause participants to react in a way that affects their behavior (p.3). In an effort to minimize the effect of my presence, I implemented an observation strategy in which participants were fully aware that they were being studied. This approach is referred to as 80 'overt' observation; in overt observation, the level of participation by the researcher may vary from one of participant -as-observer whereby the researcher assumes a dual role of worker/observer, to observer-participant involving minimal participation, to one of complete observer (Emerson, Fretz, & Shaw, 1995). As a complete observer, the researcher assumes a passive, objective stance in their observation of others. Emerson, Fretz, and Shaw argued that it is not possible for the researcher to be completely neutral, detached, and independent of the observed phenomena. This position is consistent with symbolic interactionism, which suggests that the presence of the researcher influences both the context and the meaning of the action and interactions of those observed (Crooks, 2001). In spite of this obvious disadvantage, May and Pope (1995) argued that overt observation provides an important advantage over interview alone in terms of "circumvent [ing] the biases inherent in the accounts people give of their actions caused by factors such as the wish to present themselves in a good light, differences in recall, selectivity, and the influences of the roles they occupy" (p. 183). I assumed an active participant role. As an active participant, I remained visible to the participants but refrained from participating in any physical contact or interactive activities such as interviewing, interjecting, or seeking clarification during the observational experience. Symbolic interactionism supports my assumption that my presence affected the behavior of those being observed. In this sense, I was an active participant. Following any observational experience, I tried to interview participants to understand how they perceived my presence and how this influenced their actions or their perceptions of the actions of others. Post-observation interviews were not always possible. For example, often when day-to-day interactions or other activities occurred while I was on the nursing unit, participants in observed interactions or activities were not interested or were too busy to be interviewed. 81 While observation in the field allows the researcher to collect data related to human activity and the environment that interview data alone do not provide (Morse & Field, 1996), I had to consider informed consent. The main concern related to observation is to ensure subjects are protected from harm. Therefore, prior approval of an observational guide was obtained as part of the ethical review process. The observation guide (Appendix M) identified with as much predictability as possible who and what I intended to observe. My observations were focused on verbal and non-verbal communication between participants, physical characteristics of the participants, and the nature of the environment. The guide was not static; it became more refined as dictated by developing theory. Numerous participant observational experiences were planned and/or arose spontaneously while in the field. Since the majority of participant interviews were conducted in the acute care hospital setting, many interactive episodes between various nurses (and other health care professionals), patients, and/or family members occurred in my presence. Such interactions provided important opportunities to explore participants' perspectives about the nature and frequency of such interactions, how interactions may have been affected by the presence of the researcher, as well as the meaning and significance of such interactions. For example, during an interview with a family member participant (F-009) that occurred at the bedside, the nurse poked her head through the curtain drawn around the bed and began an interaction with the family member related to the care and progress of her mother (the patient). On the basis of this observation the following participant observation note was written: Prior to turning the recorder back on, I asked the family member: From your perception, how meaningful is the exchange between yourself and the nurse that just occurred? She responded by saying that it was very important to her that the nurses' initiate interactions with family members and that they do that by "checking on them just like L (nurse) did. She went on to say that "L" does that all the time but not all of the nurses did. Furthermore, she indicated that "working closely with the girls [the nurses] is great because then you 82 have a contact person if you need something. From this interaction on reflection, it was obvious to me that there was a level of familiarity between the nurse and the family member that allowed them to converse on an equal level with regard to the care of the patient. This observation was consistent with what the family member had described in terms of how important it was to the nurses that she has been there so much to help with the care of the patient and to act as a source of information. There was a degree of trust between them that was palpable in the exchange of information and the sense of agreement between them that everything was okay for the moment. It was interesting that once again the nurse did not acknowledge my presence at least not until the family member drew the nurse's attention to what we were doing. Because I had nothing really to do with the patient, my presence and purpose of being there seemed to me to be inconsequential from the nurse's perspectives (Participant observation note dated May 31, 2005). Several hours were also spent prior to, between, and after interviews simply observing the environment on the various acute care units in the participating hospitals. This provided me with opportunities to make general observations related to nurses' work, the systemic nature and physical characteristics of the work environment, as well as interactions between and among nurses, nurses and other members of the health care team, for example, physicians, pharmacists, unit clerks, and finally between nurses, patients and family members. In total, thirty-three hours of participant observation were completed. 4.5.8 Field notes. Since it is difficult for any researcher to remember all the details of an observational experience, field notes are an important vehicle for recounting the experience during, and more importantly, immediately following the observation (Emerson, Fretz, & Shaw, 1995). Field notes include 'jottings of salient points' or details about the environment such as sights, smells, and sounds as well as physical and behavioral characteristics of people in the setting (Morse & Field, 1995). During my observations I made notes that were reworked in detail later. I used these notes to supplement my observational experiences (Emerson, Fretz, & Shaw). I also included key events or things that stood out and my emotional, cognitive, or other personal responses to these incidents. I avoided judging the events or incidents and particularly 83 the observed behavior of others. Emerson, Fretz, and Shaw (1995) argued that "Prejudging incidents in outsiders' terms makes it difficult to cultivate empathetic understanding and to discover what importance local people give to them" (p.27). I was open to the fact that people may respond to like or similar events in contrasting ways and tried to understand why they acted in a particular way. For this reason, following Emerson, Fretz, and Shaw's recommendations, I tried to talk to the participants about their impressions. As with the act of observing others, writing notes during participant observation also requires informed consent from the participants. Furthermore, sensitivity is required in relation to when, how, and what the researcher writes in the presence of participants (Morse & Field, 1995). Participants may feel uncomfortable with or distracted by what the researcher is writing about them. Researchers should remain responsive to the feelings of others and avoid making notes about matters that may be embarrassing or confidential in nature (Emerson, Fretz, & Shaw, 1995). In an effort to capture detailed description of the interactive processes, it is important that field notes be written as soon as possible following the observational experience (Emerson, Fretz & Shaw; Morse & Field). During the course of data collection, I informed participants prior to the start of each interview that brief notes might be taken in an effort to capture main themes of the conversation and/or ideas about subject areas requiring clarification or elaboration. Despite the need to do this, I made every effort to focus primarily on the participant and what they were saying. Following each interview and observational experience, field notes were written as soon as possible (Appendix N- Sample field notes). Although it was difficult, I wanted to capture as much objective observational detail about the subjects and the environment as possible. I offer this excerpt from the field notes written following patient interview 05 as an example: 84 We introduced ourselves to each other and both sat in chairs on opposite sides of a small desk in the Patient Care Coordinator's office. The office was small and had no windows. The participant positioned a portable oxygen tank beside her and adjusted the oxygen tubing and nasal prongs that she was wearing. Her breathing was labored; she exhaled noisily through pursed lips. She was pale and slightly hunched over. Her attire was casual 'street clothes' - sweat shirt and bottoms. Before leaving us, the PCC told the patient to let her know if she needed anything and reassured her that I wouldn't make her nervous. Earlier I had been told by the PCC that the patient's breathing tends to become more labored with increased anxiety. The patient stated that she "would be fine". The door was closed so we had complete privacy and there were no interruptions at any time during the interview. Throughout the interview I found her [the patient] to be very articulate, despite her labored breathing. She explained to me that she was doing "controlled breathing" - slow, steady breaths in through her nose and out through her mouth. She took her time answering questions (due to the difficulty she experienced breathing and talking at the same time) but she had a good level of understanding; took time to critically reflect on the questions before answering, and did not hesitate to ask for clarification if she did not understand a question posed to her (P-05, September 9, 2004, p. 1). While maintaining objectivity is important when writing field notes, it is equally important to keep track of subjective biases, hunches, and hypotheses in relation to the setting or phenomenon in a personal journal or diary (Morse & Field, 1995). Richardson (2000) referred to this form of field note as 'personal notes' (p. 941). Richardson argued that it is important to record feelings and impressions because they affect what and how we claim to know. 4.5.9 Personal journal. Throughout the process of data collection, I became aware of subjective, personal biases or unsubstantiated hunches related to the substantive area, setting, and phenomena. Although my hunches may have later proved erroneous, Morse and Field (1995) maintained that it is important to keep track of subjective impressions. Hall and Callery (2001) argued that it is important that the researcher make transparent the influences of investigator-participant interactions (reflexivity) and power and trust relationships (relationality) during the research process in order to enhance the validity of the findings. Journaling provides a vehicle by which the researcher can 85 explore and explain his/her own perceptions and constructions of the phenomenon and acknowledge how these affected the inquiry (Chiovitti & Piran, 2003). Throughout the study, I recorded my personal reflections on interview or observational data. My writing generally included reflections on the effect that I may have had on the participant(s) or on our overall interactive processes, specifically with regard to issues of power and trust. For example, in addition to a field note written following nurse participant interview 03,1 made a separate journal entry to record my thoughts about how my possibly overly enthusiastic verbal and non-verbal responses to the nurse's perspective on family may have influenced how she answered my questions. The following is an excerpt from my journal entry dated August 04, 2004: I was feeling very much at ease with the nurse and she appeared relaxed with me and with the interview process. I found myself especially keen to hear her thoughts about how the family theory that she had been taught didn't quite fit in with practice in the ICU and probably probed a little too much around this topic. I know that I showed approval for her comments and prompted her to continue and to elaborate during this part of the interview. It was likely that she was encouraged to continue based on my positive response but is it possible that I was overly encouraging? I will have to be aware of my verbal and non-verbal responses to comments made by participants. Giving off messages of approval or disapproval may potentially influence their perspective more so than is desirable to maintain openness to participant's views. Reflecting on this entry provided an opportunity to think carefully about how my responses could influence participants and the data. I realized that messages of approval or disapproval were forms of power that could influence how participants responded to my questions. The insight gained through journaling prompted me to take greater care to be aware of my verbal and particularly my non-verbal responses during interviews. 4.6 Data Analysis From the beginning of data collection, I engaged in constant comparative analysis or 86 generating theory from the raw data. Glaser and Strauss (1967) described the constant comparative method as "generating and plausibly suggesting...many categories, properties, and hypotheses about general problems" (p. 104). The constant comparative method, which involves comparing and contrasting data by incident, clustering codes to develop categories, integrating categories and their properties, delimiting the theory, and finally, writing the theory, required me to use a balance of inductive and deductive logic. Inductive logic was reflected in the ongoing construction of theory from data; direction for further data collection was an outcome of deductive analysis of codes induced from data. One outcome of concurrent data collection and analysis, therefore, was my increased sensitivity and wisdom about my data in terms of "where to take it conceptually and where to collect more data" (Glaser, 1978, p. 6). Central to the process of data analysis in grounded theory is the conceptual code. Coding provides the basis for interacting with and naming and categorizing data, and occurs on a continual basis as data are collected and studied (Charmaz, 2000). As Glaser maintained, coding ones data allows the investigator to move to a theoretical level of analysis by fracturing data and grouping them into codes that conceptualize underlying patterns within the data (Glaser, 1978). By developing relationships between conceptual codes, one develops grounded theory (Glaser). Two general types of codes are generated: substantive codes and theoretical codes (Glaser, 1978). I used substantive codes to conceptualize the "empirical substance of the area of research", whereas I used theoretical codes to "conceptualize how the substantive codes may relate to each other as hypotheses to be integrated into the theory" (Glaser, p. 55). I began coding with open coding followed by selective coding for a core variable or variables. 87 4.6.1 Open coding. Open coding was described by Glaser (1978) as "coding the data in every way possible" (p. 56). Open coding procedures involve the initial development, comparison, and contrast of incidents within the data to identify as many categories as possible (Dey, 1999; Glaser 1978; 1992). As coding progresses, new categories are constructed and new incidents are identified that fit under existing categories. Glaser (1978) maintained that the process of open coding allowed the analyst to "see the direction in which to take his [her] study by theoretical sampling before he [she] becomes selective or focused on a particular problem" (p. 56). To ensure that open coding procedures are used properly, Glaser (1978) offered a set of general rules to govern the process. The first rule was that the researcher must constantly ask three questions of the data: "What is the data a study of? "What category does the incident indicate"? What is actually happening in the data or what is the basic social- psychological problem that the participants face? Continually asking these questions enables researchers to be sensitive to new issues and focus on patterns among incidents that give rise to codes (Crooks, 2001; Glaser). The second rule is that open coding requires line-by-line analysis. The third rule is that the analyst must do his/her own coding. While conducting line by line analysis, I assigned codes or descriptive labels to phrases, sentences, or groups of sentences within the data. Initially, I labeled incidents on the basis of "in vivo" codes, meaning in "the language of the substantive data itself (Glaser, 1992, p. 45) and expressed them as gerunds. Gerunds or action words are used as much as possible to preserve the emic perspective (Morse & Field, 1995). For example, open (in vivo) codes identified in the first interview conducted for this study (P-01) included: feeling vulnerable, complying, feeling reassured, just floating, [nurses] just doing their job, telling, 88 checking on, assuming. For Glaser (1978), line by line coding of all data was important in that it "forces the analyst to verify and saturate categories, minimizes missing an important category, produces a dense rich theory, and gives a feeling that nothing has been left out" (p. 58). I found that coding the data myself was not only efficient but also enhanced my theoretical sensitivity which is important to achieving conceptual saturation, as suggested by Glaser. As more data were collected, I created new categories and new incidents fit existing categories until patterns began to repeat. For example, while conducting line by line coding in the context of Glaser's three questions for the first interview conducted (P-01) the following ten codes were generated: getting to know, asking questions, busyness/time, expectations, information sharing leads to trust, impact of social forces, going along with it, power and control of information, what nurses do, and personality factors. For subsequent data, I not only engaged in line by line coding, but also comparison of codes. By the fourth interview (P-03), additional codes such as 'intelligent selection' which seemed to describe a process of relationship management that was nurse driven, and 'patient vulnerability' or the need to feel secure were created. By the eighth interview the notion of the squeaky wheel had been identified by several participants. As data collection and analysis progressed, I saw incidents in the data that seemed to be related. For example, codes such as information sharing, developing trust, and patient vulnerability seemed to emerge as a significant component of a higher level category referred to as 'building rapport'. Following interviews N-03, N-04, and N-05 I identified the concept of 'stepping back' or moving to a lower level of engaging. I also identified a clearer pattern of interaction among nurses, patients, and family members that not only included various 89 progressive stages of development, but also conditions and barriers that constrained or facilitated the process. Although it was ultimately my interpretation of the data that shaped the emerging codes and categories, Glaser's (1992) third rule, which relates to theoretical sensitivity and is consistent with symbolic interactionism, is that I must remain open to the data and avoid a priori hypotheses. In other words, I must seek meaning and understanding from the actors' perspectives as to the factors that influence how they behave in a given situation, and what meaning they attach to their actions. In Glaser's view, remaining open to the data means that the "analyst starts with conceptual nothing - no concepts" (p. 39). As I engaged in the coding process, I experienced insights, feeling, thoughts, and later ideas about emerging relationships and captured them in memos. Glaser's fourth rule, which is critical to the process of generating grounded theory, is to always interrupt coding to memo ideas. Glaser (1978) wrote: "memos are the theorizing write-up of ideas about codes and their relationships as they strike the analyst while coding" (p. 83). Memoing is a constant process that serves several important functions in generating theory (Glaser). For example, memo writing helps track and preserve ideas; promotes insight into tacit assumptions; increases the conceptual level by identifying themes and patterns in data; captures speculations about the properties of the categories, relationships between categories, or possible criteria for selection of additional participants; and facilitates the integration of categories to generate theory (Glaser; Morse & Field, 1995). To illustrate the importance of memos to the process of generating theory that is grounded in the data, I refer to a series of memos I wrote related to the notion of collaboration - a concept I thought to be relevant in the early stages of data collection. 90 March 2005: In acute care hospital settings, nurses do not necessarily collaborate with all patients and/or family members.. .The primary reason cited for this is because nurses either 'do not have time' or do not perceive that there is a need. Instead, collaborative episodes seem to occur primarily on an 'as needed basis'. April 2005: It is becoming more and more evident that collaborative processes between nurses, patients and family members occur on a selective basis. That is to say that nurses appear to collaborate with patients and/or family members in different ways, for different reasons, and for varying amounts of time... July 2005: In acute care hospital settings, nurses do not necessarily collaborate with all patients and/or their families - this has been consistently acknowledged by all nurse, patient, and family member participants. In fact, the data suggest that there are instances when nurses, patients, and especially family members do not interact at all - at least not in a therapeutic way. December 2005: .. .various levels of interaction between nurses, patients and/or family members can be identified within the context of the acute care hospital environment. The initial level of nurse, patient and/or family member interaction achieved is dependent on the extent to which various systemic and contextual conditions align to create an environment conducive to interaction. The more mutually favorable terms or conditions are perceived to be, the more quickly nurses, patients and/or family members move towards achieving personal engagement.. .There can also be a decline in the level of engagement. As demonstrated by these memos, although I initially thought collaboration might be a theoretically significant category to explain how nurses, patients, and family members interacted, with further sampling and analysis, it became evident that collaboration did not capture the nature of interaction described by the participants. Instead, a category I referred to as engagement or engaging emerged. In response, I resumed collecting data through theoretical sampling and began delimiting the theory through selective coding to determine if progressive engagement 'best fit' the pattern of interaction reflected in the data (Glaser, 1978). This recursive process was consistent with the grounded theory method, where tentative theoretical explanations are generated and modified on the basis of incoming data that confirm or refute these explanations (Davis & Harris, 1989; Sandelowski, 1995). This process allows the analyst to become increasingly sensitized to categories that may provide an answer (Davis & Harris; Glaser; Sandelowski). Once the analyst begins to see the prospects for a theory, he/she begins selective coding (Glaser). 4.6.2 Selective coding for a core category. Following my hypothesis that progressive engagement met the criteria for a core category, I undertook a recursive process of theoretical sampling, constant comparison of data sets and selective coding over several months. Selective coding refers to the process of categorizing, re-categorizing, and condensing all first level codes around a possible core category in an effort to determine if it accounts for most of the variation of the central phenomenon of concern and integrates all other categories (Glaser, 1978; Kendall, 1999). I used second level coding to contribute to raising the conceptual process to a higher level of abstraction to account for more variability in the data, as well as facilitate theoretical sampling to delimit the theory to one core variable as suggested by Glaser. Upon commencing second level coding, I theoretically sampled around concepts that seemed related to progressively engaging. For example, I developed codes such as the squeaky wheel, taking time, stepping back, and feeling each other out and developed conditions that seemed to relate to hypothesized stages for engaging. When I had completed forty interviews, I found that my tentative core category - progressively engaging held a central position to other possible categories and through selective coding it became clear to me that this proposed core category was "processing out" (Glaser, 1978). In other words, for me it appeared increasingly evident that progressively engaging was not only a core category but also a process which accounted for two or possibly three stages of engagement. Each hypothesized stage seemed to account for various factors such as acuity (e.g. medical diagnosis, nursing diagnoses, communication of need), time (e.g. acuity, priority, work load, competing demands), personal factors (e.g. language barriers, personality, the 'squeaky wheel'), and systemic factors (e.g. rules, routine, lack of privacy); getting to know; and/or building rapport. Through one-upping with my supervisor (Glaser, 92 1978) it became evident that there were conditions or terms of engagement that determined the stage of engagement attained in relationship as well as conditions in which nurses, patients and/or family members stepped back from engaging or failed to engage. 4.6.3 Core category. Glaser (1978) maintains that the goal of grounded theory is to generate a theory that accounts for how participants process the primary concern or problem in the setting. The core category is central to constructing theory because it accounts for most of the variation in a pattern of behavior (Glaser); all other categories and their properties are related to it. Based on its ability to integrate categories, the core category leads to theoretical completeness but remains amenable to modification in response to changing conditions (Glaser) Identifying a core category takes time; however, it eventually stands out on the basis of its ease of saturation, relevance, and workability in relation to all other categories (Glaser). Additional criteria for determining the core category include: its frequent reoccurrence in the data; it's clear and grabbing implication for formal theory; and the relevance and explanatory power it provides in the analysis of the processes under consideration (Glaser). As my ongoing search for a core category unfolded, possible core categories such as selective collaboration that I believed offered an explanation of what was happening in the data ultimately failed to "carry through" (Glaser, 1978). In other words, the notion of selective collaboration could not fully account for the pattern of interaction that emerged as data collection and analysis progressed. Instead, a new tentative explanation in the form of a category referred to as progressively engaging was created. Progressively engaging referred to a continuum of interaction. My early conceptualization of this continuum included being less engaged (referred to as 'just doing the job) at one end to being fully or more personally engaged (referred to as 93 'doing the job with heart') at the other. Moreover, the category of progressively engaging seemed to account for particular conditions that participants described that either facilitated (e.g. personality, increased acuity) or constrained (e.g. perceptions of time, language barriers) efforts to manage nurse-patient and/or family relationships in the acute care hospital settings by engaging. In other words, there seemed to be conditions that influenced nurses', patients' and family members' willingness to engage with each other at various stages on the continuum. Early conceptualization of the core category of progressively engaging was captured in the memo excerpt that follows: The continuum of interaction that reflects the level of nurse, patient and/or family interaction in acute care settings at any given time (at least those included in this study), appears to be characterized by obligatory levels of interaction at one end of the continuum through to progressively higher levels of engagement to possibly achievement of personal engagement at the other end. Interaction characterized as being obligatory in nature is associated with what nurse, patient and family member participants consistently describe as 'just doing the job'. Nurse, patient and family member participant descriptions of nursing care delivered at the level of 'just doing the job' is focusing on the elements of nursing work without engaging with others on a personal level (December, 2005). At this earlier stage of theory construction, I believed the core category represented a basic social psychological process, because it captured other categories that were suggestive of progressive stages of engagement. The categories that related to the core were establishing competence, professional engagement, and personal engagement. On the basis of considerably more analysis via concurrent processes of coding, one-upping with my supervisor, theoretical and process memoing, and constant comparison, a substantive theory was drafted - although several revisions to the theory were undertaken (Appendix P). As ongoing analysis enhanced my levels of conceptual abstraction and I accounted for more variation in the data, I next determined how the core category and the substantive codes were theoretically related. 4.6.4 Theoretical coding and sorting. In an effort to confirm my hypothesis that progressively engaging was a basic social psychological process that explained how nurses, patients, and family members managed their relationships in acute care hospital settings, I conducted theoretical coding. When theoretically coding, analysts systematically relate categories in theory by "recognizing what is important in the data and giving it meaning" (Morse & Field, 1995, p. 161). According to Glaser (1978), "theoretical codes conceptualize how the substantive codes may relate to each other as hypotheses to be integrated into a theory" (p. 72). Theoretical coding promotes further conceptual abstraction and generates new ideas that give the theory integrative scope, breadth and perspective (Glaser). To enhance the process of theoretical coding I constructed a theoretical diagram (Glaser, 1978). Developing a diagram is a way to theoretically code pictorially, which facilitates writing up hypotheses about the substantive meaning of the connections depicted (Glaser). The diagram was useful as a basis for discussion between me and my supervisor - discussions which facilitated the achievement of continually more advanced conceptualizations of the theory. Among these advanced conceptualizations was my construction of the theoretical code 'terms of engagement'. Initially terms of engagement accounted for those conditions that facilitated a willingness to engage at the first stage (focusing on tasks) but with further analysis the terms of engagement came into play at each stage of the process. I undertook several revisions to my diagram to reflect these theoretical advances. I undertook theoretical or conceptual sorting concurrently with theoretical coding; the diagram was extremely useful for the sorting process. Theoretical sorting refers to "putting the fractured data back together" but on a conceptual, creative level (Glaser, 1978). It is a very 95 important step preceding the writing of the theory because it forces connections between categories and properties, generates dense, complex theory by stimulating more memos, and eventually integrates relevant literature into the theory (Glaser). The process of theoretical sorting (and coding) was facilitated by my theoretical sensitivity. Glaser (1992) described theoretical sensitivity as being dependent on the researcher's knowledge, understanding, and skill, and which in turn "fosterfs] his [her] generation of categories and properties and increases his [her] ability to relate them through hypotheses, and further integrate the hypotheses, according to emergent theoretical codes" (p. 27). I used my knowledge about family nursing, skill at data analysis, and intense interaction with the data to develop concepts and relate them to one another. Simultaneously, I tried to treat any predetermined ideas as simply another source of data. The core category 'progressively engaging' related easily and meaningfully with the other categories constructed, namely: focusing on tasks, feeling each other out, building rapport, stepping back, opting out and falling through the cracks. While core categories can take many forms (e.g. a process, a condition, a consequence), if it is a process it must have at least "two or more clear emergent stages" that reflect a process by which participants manage the problem under investigation and is hence referred to as a Basic Social Psychological Process (BSPP), (Glaser, 1978). Glaser (1992) defined basic social processes as "fundamental patterned processes in the organization of social behaviors which occur over time and go on irrespective of the conditional variation of place" (p. 100). 4.6.3 Saturation and completeness. When second level coding of all sources of data (memos, field notes, transcripts) no longer yields new information, that is, new data fit into existing categories (Charmaz, 2000), and the 96 core category accounts for most of the variation in behaviors, saturation is reached (Morse & Field, 1995). In the context of saturation, Glaser (1978) referred to cutting off rules which include running out of memos, as well as theoretical and scholarly completeness. Theoretical completeness means that the analyst has explained the phenomenon under investigation with as few concepts as possible and with as much scope and variation as possible (Glaser). Based on the analyst's knowledge of the literature, scholarly completeness refers to the point in which the analyst feels that his/her theory makes a "integrative and recognitive" contribution to the relevant literature (Glaser, p. 126). Saturation and the subsequent cutting off of theory construction is, therefore, a subjective judgment - a conscious decision by the analyst based on the belief that he/she can't get any more from his/her 'conceptual work (Glaser, 1978). After analyzing forty participant interviews, the patterns of behavior that emerged were accounted for in categories I constructed which captured the stages of progressively engaging. For example, focusing on tasks (the first stage of progressively engaging) accounted for when nurses and patients and/or family members came together - that period of time when there was limited familiarity with each other and the primary focus of interaction was on the delivery of priority physical/medically delegated care. Given the constraints of dissertation research, such as time and resources, I, in consultation with my research supervisor, concluded that my conceptualization of progressively engaging, including the three stages of engaging, met the criteria for saturation as described above. It explained how nurses, patients, and family members managed their relationships within the acute care hospital settings and accounted for the conditions or factors that participants described that facilitated or constrained their relationships with one another. Moreover, I found that progressively engaging made a significant contribution to the substantive literature by offering an explanation of nurse, patient, family interaction that is grounded in practice - that is, it is based on the perspective of participants actively engaged in the social psychological process of managing their relationships in these settings. One of the most significant tests of saturation however, is that the conceptual work meets the requirements for rigor - that is, the theory constructed sufficiently explains with "concepts that fit, work, have relevance and are saturated" (Glaser, 1978, p. 125). 4.7 Criteria for Rigor Theory, according to Glaser and Strauss (1967), is a strategy for handling data by providing a conceptual framework for describing and explaining it. Theory must be readily understandable, relevant, have practical application to the situation under study, and "provide clear enough categories and hypotheses so that crucial ones can be verified in present and future research..." (Glaser & Strauss, p. 3). Glaser and Strauss maintained that the best approach to generating valid social theory is through systematically developing theory from data. Indeed, when I used the grounded theory process with its guidelines and detailed strategies for collecting, coding, analyzing, and presenting data, it went hand in hand with verifying it [emerging theory] (Glaser & Strauss, p. 3). It is on the basis of how the theory was generated, as well as how much 'grab' it has, that the validity of grounded theory ought to be judged (Glaser, 1978; Glaser & Strauss; Lomborg & Kirkevold, 2003). The term 'grab' is used by Glaser and Strauss (1967) and Glaser (1978; 1992) to describe grounded theory deemed to be well constructed, because people find it interesting, meaningful, and reflective of the essence of the phenomenon under investigation. In order to attain grab, however, the theory must meet certain criteria. These criteria include: fit, work, relevance and modifiability. The notion of 'fit', initially described by Glaser and Strauss, means that the theory constructed from the data "fits the situation being researched" (1967, p. 3). In other words, the theory clearly corresponds with the social reality and is; therefore, readily understandable to those for whom it is relevant. In order to achieve fit, a theory must account for most of the variation in the data, codes, and categories. In other words, the most salient codes and categories account for the greatest amount of variation in the data. Categories must be "readily applicable to and indicated by the data under study" as opposed to forced into pre-conceived or pre-existing theoretical categories (Glaser & Strauss, p. 3). According to Glaser (1978), an important property of the notion of 'fit' is emergent fit. Emergent fit involves concurrent refitting of extant categories with emerging data to ensure that the category still 'works', thereby earning its way into the emerging theory. Allowing for changes from emerging data contributes to the modifiability of the theory. I attended to the criteria of emergent fit throughout this study by employing the peer research support group strategy. This strategy enhances the quality of qualitative research by facilitating the process of review and commentary by support group members regarding such documents as interview transcripts and memos, and provides a forum to discuss the researcher's ideas (Jacelon & O'Dell, 2005). Specifically, I sought feedback from my research supervisor. I did this by submitting written memos, copies of transcripts, proposed substantive theories, and draft descriptions of categories and the core variable at various stages of their development to my research supervisor for feedback about how well the emerging categories fit the data. I also met with my supervisor regularly throughout the study (and with the other members of my research committee later in the process), to discuss my data analysis work; obtain feedback and exchange ideas. In addition, I took advantage of opportunities to discuss my proposed substantive theory at 99 various stages of its evolution with colleagues and peers. For example, as a guest speaker in a graduate level qualitative research course I presented an early version of the core category, progressively engagement, and received feedback from graduate nursing students. Based on their knowledge of the substantive area and their clinical knowledge and practice, they indicated that the core category and stages resonated with their experiences and expressed enthusiasm for the ability of the theory to explain interpersonal dynamics associated with nurse, patient, and family relationships in acute care hospital settings. As previously mentioned, grounded theory must also 'work'. Glaser (1978) described the notion of work as: "a theory should be able to explain what happened, predict what will happen and interpret what is happening in an area of substantive or formal inquiry" (p. 4). In order for theory to 'work' Glaser emphasized the need to "get the facts" about what is "really going on" in the substantive area (p. 4). While obtaining the perspective of the actors is critical to this aim, the underlying assumption is that a natural world is available for observation and analysis (Hall & Callery, 2001, p. 260). This becomes problematic especially if one adopts an interpretive stance such as symbolic interactionism which assumes that meaning is created through interaction. Hall and Callery wrote: Because data are produced through the criterion of meanings during processes of interaction, it follows that the quality of the data will be influenced by the nature of the relationship between the researcher and the participant, therefore, the nature of that relationship requires the attention of the grounded theory investigators (p. 260). Hall and Callery further argued that grounded theory researchers must take measures to account for the effects of subjectivity on the research process in order to enhance the validity of the emerging theory. Such measures include the incorporation of reflexivity which "addresses the influence of the investigator-participant interactions on the research process", and relationality which "address power and trust relationships between participants and researcher" (p. 258). 100 In this dissertation study, I dealt with reflexivity and relationality by utilizing personal journaling as a self-monitoring tool. Post-interview and observational journaling provided a vehicle for tracking and increasing my sensitivity to personal views and constructions and how these might influence the inquiry process (Chiovitti & Piran, 2003). For example, following my interview with a patient and her daughter the following entry was made in the researcher's journal: While both mother (patient) and daughter describe a process whereby some nurses went more out of their way [e.g. showing interest and concern] than others, the patient's view seemed to be tempered by her experience in hospital e.g. watching nurses cope with the many demands in their work" (June 30, 2004). This insight contributed to the level of theoretical sensitivity in relation to patients' perceptions of contextual and systemic factors vis-a-vis their interactions with nurses as opposed to family members who may not have the same awareness of such factors. In another example, I addressed relationality in the following journal entry dated June 30, 2004: I found myself feeling frustrated with what I perceived to be the lack of honesty by this nurse in terms of how she portrayed her interactions with family. It just seemed so inconsistent with what I have heard from others and from what I have read that it just seemed to good to be true! I was kinda aware of my bias (or possible bias) but pressed on anyway trying to get at 'what really happens' as opposed to what the RN wanted me to think happens. Because the interview happened at the spur of the moment and because we didn't really have any time to get comfortable with each other, I wonder if the nurse felt threatened by me and by my questions. She may have been skeptical about my intentions and maybe she thought I was 'investigating' how well the nurses on her unit did family nursing - or not. She may have felt on the defensive a bit. This journal entry provided the opportunity to reflect on a possible power imbalance between myself and a participant. It became evident to me that not only could this imbalance affect the data collected, but it could also influence how I interpreted the data. This realization resulted in modifications in my approach, including making sure that there was sufficient 101 time to conduct interviews without being rushed and incorporating more opportunity for participants to ask me questions about myself and/or the research I was conducting. It also was an opportunity to reflect on the importance of journaling about my impressions in order to increase my awareness of how they influenced data interpretation. Returning to Glaser and Strauss' (1967) criteria of rigor, when grounded theory fits and works, relevance is attained. Relevance means that the core basic social problems and the basic social processes that have emerged are recognizable, comprehensible, and plausible to others including the participants and are parsimonious, yet broad in scope (Crooks, 2001; Glaser, 1992; Morse & Field, 1995). Parsimony and scope refer to the ability of the theory to account for as much variation in behavior within a substantive area as possible. Breadth of the theory refers to its ability to be 'sufficiently general' meaning that it is "applicable to a multitude of diverse daily situations as they change through time" (Glaser & Strauss, p. 237). I was able to account for the criteria for relevance in my developing theory by employing a form of member checking (Sandelowski, 1993). Member checking involves a process of seeking feedback from study participants about how they perceived the relevance and meaning of the researcher's development of categories that seemed to capture participants' descriptions of their experience related to nurse, patient, and family member interactions. For example, in an effort to understand and verify the relevance of how the social structural condition of asking questions related to the category of nurse busyness, and influenced nurse, patient, family member interactions, I asked nurse participant 16 the following question: Researcher: I just want to ask you about questions. I am hearing a lot about this notion of questions and because nurses are so busy um, that a lot of times there is an expectation by nurses that if people need something or you, a patient or a family member, they will ask. Does that resonate with you or no? Nurse: Not always. Some of them say, you know, well I didn't ask because I see that you are so busy that I hate to ask. (May 4,2005, N-16, 347-357). 102 In another example, I sought to determine the relevance and meaning of the emerging category of nurse busyness from the perspective of a family member participant when I asked: I have talked to a lot of nurses, and patients, and family members like yourself about this process [of nurse, patient, family interaction] and there are some things that consistently come up and I'll throw them at you and see what you think. One of the things that comes up a lot is time - that nurses are so busy that they, it is very difficult for them to interact with families, to work with them and include them in care. Is that.. .do you see that? Do you see that nurses are too busy? Or is that not a factor? Family member: Uh, okay, I will see some nurses really busy. They doing every detail work and... But I cannot say all of them like that....etc. (June 24, 2005, F-010, 235-261). By checking the degree of recognizability, comprehensibility, and plausibility of emerging categories with participants - especially in the later stages of data collection I was able to confirm the relevance of key concepts and hypotheses about how concepts were related to one another with the participants. Also, notwithstanding time constraints and other limitations to the research study, I was also able to enhance the ability of the emergent theory to account for as much variation in behavior associated with the core category as possible. Finally, grounded theory must be modifiable. Indeed, theory generation is considered an ongoing, modifiable process (Glaser, 1978). Modifiability incorporates the recognition that the process of theory development is tentative, dynamic, and requires the researcher to be open and inquisitive with regard to new data and insights. Although the basic social processes that emerge through grounded theory remain in general, "their variation and relevance is ever changing in our world" (Glaser, 1992, p. 5). Thus, as new data emerge, the theory must be modified to fit these variations. In this way, grounded theory maintains relevance and parsimony in terms of its explanatory powers. As circumstances change and new data surface, the theory should be amenable to modification in order to account for them. 103 Throughout the process of developing my theory of progressively engaging several modifications to the theory were necessary in response to the emergence of new data and insights. The concept of positive and negative interpersonal dynamics, for example, was initially conceptualized as factors affecting how nurses, patients, and family members first came together around care. With further analysis, however, I determined that interpersonal dynamics influence nurse-patient and/or family relationships at each stage of progressively engaging. Consequently the emerging theory was modified to fit this new insight thereby enhancing its relevance and workability visa vie the data. 4.8 Limitations Through use of the qualitative research process, the researcher strives to make sense of reality by examining phenomenon for patterns in meaning and to infer more abstract generalizations in the form of concepts and relationships between concepts - in other words the "construction" of theory (Morse & Field, 1995). Using the grounded theory method, the investigator employs a primarily inductive approach to data analysis of "everyday behaviors and organizational patterns to generate a theoretical explanation" about a specific social phenomenon (Hutchinson & Wilson, 1993). While the ultimate goal is to explain the social world through theory based on rich description of phenomenon, the level or degree of explanatory power varies depending on such factors as the nature of the research and the research question(s), sample size, length of time in the field, and so on (Hutchinson & Wilson). Since the research questions proposed in this study focused on nursing concepts grounded in a particular nursing practice context, a limitation of the study is that the theory constructed is applicable only to similar groups (nurses, patients and family members) in similar contexts and, therefore, constitutes a mid-range theory (Morse & Field, 1995). The research findings are also 104 constrained by the small sample size secondary to the nature of the busy acute care hospital setting and the degree of willingness and/or availability of nurses, patients, and family members to participate. Moreover, sample size was further constrained by such issues as limited time in the field secondary to restricted access within the hospital and Ph.D. degree completion deadlines. In addition to sample size and time limitations, the research findings reported are also constrained by age, gender, and, to a certain extent, cultural representation. The majority of patient and family member participants were Canadian, Caucasian females over the age of fifty years. Nurse participants were Canadian female registered nurses working in acute care hospital settings. Thus the theory is relevant to Canadian registered nurses working in acute care hospital settings and mid-life or elderly patients and families, but may not be applicable in other countries, to health care settings in or outside of Canada, or for younger patient and/or family populations. Because my sample is part of a larger population, some applicability may be construed (Dempsey & Dempsey, 2000). Morse (1999) argued that, although sample sizes in qualitative research tend to be smaller, the theory tends to be comprehensive, complete, saturated, and accounts for negative cases because the participants are purposefully and selectively chosen based on the anticipated contribution they can make toward the emerging theory. This means that the theory may be applicable in similar situations in terms of the nature of participants and their hospital settings. 4.9 Summary I used the grounded theory method guided by symbolic interactionism as my strategy of inquiry in this study. From the symbolic interactionist perspective, the aim of grounded theory is 105 to understand and conceptualize complex interactional processes and construct theoretical explanations for the phenomenon under investigation. In the case of my study, the phenomenon under investigation was the perceptions and behaviors of nurses, patients, and family members as they managed relationships during care giving in acute care hospital settings. In this chapter, I have provided an overview of the grounded theory method and described the research design including sampling criteria and specific data collection and analysis processes. I elucidated ethical considerations and my approaches to them. I have described the construction of my core category, which is the basic social psychological process of progressively engaging. I also presented ethical considerations and issues of rigor. I have outlined criteria for rigor in grounded theory and how I attended to rigor. Progressively engaging with its three stages (focusing on tasks, getting acquainted and establishing rapport) is a core category that explains how nurses, patients and family members manage relationships in acute care hospital settings. I will present my findings in the chapter that follows. 106 CHAPTER FIVE 5 Research Findings The purpose of this chapter is to present the findings of this research study. Specifically, I will describe the substantive theory of progressively engaging that captures how nurses, patients and family members manage their relationships during acute care hospitalization. To begin, I describe the participant sample, including demographic characteristics for the nurse, patient, and family member participant groups. Next I will present an overview of the theory. A more detailed discussion about the core variable - progressively engaging, and its major sub-categories and indicators will follow. 5.1 Sample Between May 07, 2004 and June 24, 2005 a total of forty participants were recruited. Seventeen interviews were conducted with nurses, ten with family members, and thirteen with patients. Participants were recruited from a total of ten acute care nursing units across four community hospitals in the Fraser Health Authority of British Columbia. The type of nursing units included four medical units, three surgical units, two transitory care or activation units, and one community hospital intensive care unit. 5.1.1 Description of patient participants. In total, thirteen patients consented to participate in the study. As depicted in Table 1.1 -Demographic Characteristics of Patient Participants, the majority of patients, (9/13) were female. With the exception of one patient who was between thirty and thirty-nine years of age, all patients were over fifty years of age. One participant was employed; one was semi-retired; and eleven of the thirteen patient participants indicated that they did not work or were retired. Ten patient participants were married at the time of the interview. All patient participants were Patient Participants P- 01 P- 02 P- 03 P-04 P- 05 P- 06 P- 07 P- 08 P- 09 P-10 P-11 P-12 P-13 Gender: Male/Female F M F F F F F F M F M F M Age: 0-39 yrs X >50 yrs X X X X X X X X X X X X Culture: Canadian X X X X X X X X X X European X X X Marital status: Married X X X X X X X X X X Other Divorced Widow Widow # of Children 1 2 3 5 3 2 2 3 1 3 2 3 2 Education: K- gr. 8 X X G r 9 - 1 2 X X X X X HS diploma X X X College Grad. X X X Employed Yes Retired No No Retired No No No Retired No Semi Retired Retired Retired 1 s t Hosp? Yes Yes Yes No Yes Yes Yes No No No Yes Yes Yes LOHS - days 5 3 21 5 3 11 2 6 2 4 11 30 90 Diagnosis Hyst Knee sx CA stroke COPD Back pain Fall Bowel sx # hip UTI CA # hip Stroke Support Person H W H D D H H D H W H D Legend: Hyst = hysterectomy, CA = cancer, Knee Sx = knee surgery, COPD = chronic obstructive pulmonary disease UTI = urinary tract infection, Bowel SX = bowel surgery, H = husband, D = daughter, W = wife 108 parents with between one and five children; (mean = 2.5 children). Levels of education varied with five participants completing grade nine to twelve and five obtaining a high school diploma. One participant indicated that he did not attend high school while three reported having earned a college diploma. None of the participants had university degrees. The majority (10/13) of patient participants self-identified as Canadian; three identified themselves as European. Length of stay in hospital for the patients ranged from two to ninety days; the mean length of stay in days was fifteen, (although this number is skewed by one 90 day stay). The median length of stay was five days. For the majority of patients (9/13), this was their first hospitalization for the current illness episode. Considerable variation in admitting diagnoses were observed for the patient participants interviewed. Illness episodes included: hysterectomy (surgery), knee replacement surgery, bowel surgery, two cases of fractured hip, acute back pain, multiple injuries sustained in a fall, cardiovascular accident (stroke), trans-ischemic accidents (TIA's), an acute exacerbation of chronic obstructive pulmonary disease (COPD), cancer of the liver - newly diagnosed, and metastatic bone cancer - newly diagnosed, acute urinary tract infection (UTI). When asked to identify their main support person while in hospital the majority of patients (8/13) identified their spouses; four identified their daughters; and one patient identified a friend. 5.1.2 Description of nurse participants. In total, seventeen nurses consented to participate in the study. As depicted in Table 1.2-Demographic Characteristics of Nurse Participants, one hundred percent of the nurse participants were female and held a license to practice with the College of Registered Nurses of British Columbia. Fifteen of the nurses referred to themselves as a 'staff nurse', while two identified themselves as patient care coordinators. Fifteen of the seventeen nurses held a diploma Nurse Participants N-01 N-02 N-03 N-04 N-05 N-06 N-07 N-08 N-09 N-10 N-11 N-12 N-13 N-14 N-15 N-16 N-17 Age: 20-29 yrs X X X X 40-49 yrs X X X X X >50 yrs X X X X X X X X Culture: Asian X X Canadian X X X X X X X X X X Chinese X European X X X Other X Marital status: Married X X X X X X X X X X Single X X X X X Other Divorced Divorced # of Children 3 2 0 4 2 1 7 0 1 0 2 0 2 0 2 3 3 Yrs of Practice: 1 - 5 yrs X X X X 6 - 10 yrs X X 11- 20 yrs X X X X X 21-30 yrs X X X >30 yrs X X X Title P C C S N S N S N S N S N SN SN S N S N S N S N S N SN S N S N P C C Employment status FT PT PT FT FT PT PT PT FT FT FT FT PT FT FT FT FT Work hours/week 37.5 33 44 24+ 37.5 24-26 33 44 44 37.5 37.5 35-40 24-36 48 30+ 40 36 Nsg. Ed.: Diploma X X X X X X X X X X X X X X X BSN X X Specialty X X X Other X Legend: SN = Staff Nurse, PCC = Patient Care Coordinator, FT = Full time, PT = Part time © 5 © (TO 2 ET re' o a s 65 re re 83* ft' CO e Z s -! Vi re 69 re -3' 69 D O 110 in nursing education. Two nurses with diploma credentials also held a certificate in a nursing specialty or non-nursing related field. One degree nurse also reported having a certificate in a nursing specialty. A total of five nurses reported having post-secondary credentials in non-nursing fields of study. The majority of nurse participants (11/17) were employed on a full-time basis. The number of hours worked per week ranged from twenty-four hours to fifty-two hours with an average number of hours of 38.5. There was considerable variation among the nurses in terms of years of practice - between one and greater than thirty years. The majority of nurses (5/17) reported having between eleven and twenty years of experience. Three nurses reported having more than thirty years of experience. Most of the nurse participants (8/17) were over fifty years of age. The majority (10/17) of the nurses were married. The number of children among the nurse participants ranged from zero to seven (mean = 2.6). Most of the nurses (10/17) self-identified as Canadian. 5.1.3 Description of family member participants. Family members were the most challenging group to recruit, because their visits were unpredictable and infrequent in nature, and many were unwilling to participate due to lack of time or other undisclosed reasons. Ten family members were interviewed in total, as depicted in Table 1.3 - Demographic Characteristics of Family Member Participants. One hundred percent (100%) of the family members were female and included four wives, five daughters, and one sister. Each family member participant either self-identified or was identified by the patient and/or nursing staff as the primary family care giver/support person to the patient while in hospital. The majority of family participants (8/10) were over the age of fifty years and self-identified as Canadian. Seven of the ten were married. All but one family member had children. The mean number of children among the remaining nine family members was two. There was a I l l range of between one and four children. In terms of education, two family members reported having attended grades nine to twelve; two graduated from high school; three completed some college courses; three completed a college diploma. None of the family had less than a grade nine education and none had participated in university education. Six of the family members reported that they were retired or did not work. Four family members indicated that they were employed; two were employed on a part-time basis. Only one of the family members indicated Table 1.3. Demographic Characteristics of Family Member Participants Family Members F-01 F-02 F-03 F-04 F-05 F-06 F-07 F-08 F-09 F-10 Age: 20-29 yrs X 40-49 yrs X > 50 yrs X X X X X X X X Culture: Canadian X X X X X X X X Chinese X X Marital Status: Married X X X X X X X Single X Other widow Div. # of Children 2 2 2 0 1 3 2 2 3 4 Education: gr. 9 - 1 2 X X HS diploma X X College X X X College grad. X X X Employed retired Yes semi-retired part-time retired sick leave retired No Yes No Relationship to Pt. W D W D wife D S W D D LOHS of pt. in days 3 5 11 20 7 7 3 16 12 4 Pt. diagnosis knee Sx Stroke cancer cancer cancer cancer heart failure pneu-monia # hip Stroke Days missed work 0 0 0 0 0 0 0 0 14 0 Legend Div. = divorced, D - daughter, W = wife, S = sister Sx = surgery, # = fractured that she had missed time at work due to the hospitalization of her ill family member. The reasons for the ill family member's hospitalization included: total knee replacement, cardiovascular accident (CVA), cancer, pneumonia, congestive heart failure, and fractured hip. The mean length of hospital stay among the ill family members of family participants was nine days with a range of three to twenty days. Although the demographic questionnaire did not include a question about how much time family members spent at the hospital, participants did provide this data 112 while being interviewed. The amount of reported time spent at the bedside varied; however, nine of ten family member participants indicated that they visited their ill family member on a daily basis and many spent several hours each day at the hospital. 5.2 Progressively Engaging: Theoretical Overview Progressively engaging is a basic social psychological process that accounts for how nurses, patients, and family members manage their relationships around care in acute care hospital settings. Progressively engaging describes how nurses, patients, and family members 'come together' in the course of patient care with varying levels of engagement. For example, some nurses, patients, and family members described their engagement as being more formal or business-like; others experienced a deeper, more personal engagement. Levels of engagement were represented by three stages which included focusing on tasks, getting acquainted, and establishing rapport. The stages of engagement were affected by structural conditions and personal factors relevant to nurses, patients, and family members. Those conditions and factors facilitated or constrained efforts to manage their relationships by contributing to or detracting from shared meanings and perspectives. The level of engagement achieved in nurse - patient and family member relationships corresponded with feelings of satisfaction expressed by nurses, patients, and family members about their relationships and nursing care. As nurses and patients and/or family member(s) came together to manage relationships, progressively engaging reflected their descriptions of both what they perceived to be less than ideal ways of managing their relationships, as well as what they perceived as ideal ways of managing their relationships. Their perceptions were based, in part, on the nature of the interpersonal dynamics they experienced in relationship. Interpersonal dynamics characterized as 113 impersonal or business-like interactions could initiate the process of progressively engaging but also perpetuate a lower level of engagement. When nurses, patients, and family members experienced negative interactions they expressed dissatisfaction about the ways their relationships were being managed. Dissatisfaction was articulated as lack of trust and respect. For example, family members recounted feeling "invisible"; patients expressed feeling like "just another patient in the bed"; and nurses described feeling like "handmaidens". The outcomes of lower levels of engagement were described as nurses "just doing the job", wherein the focus of care was on routines and skill competencies, without a sense of connection. If interpersonal dynamics were perceived positively and fit with what nurses, patients, and family members believed ought to happen in managing their relationships, higher levels of engagement described as more personal human connections were attained. When nurses', patients' and families' efforts to manage relationships achieved more personal connections, they expressed satisfaction. Being satisfied included a heightened sense of trust and respect, acting in reciprocal ways, and better care outcomes. Patients and family members experienced a greater sense of well-being, while nurses described feeling they had done their jobs well and all groups described 'doing the job with heart'. The process of progressively engaging occurs during acute illness episodes in hospital. Because nurses provide care to all hospitalized patients, some level of engagement occurs between them and their patients in the process of providing care. In some contexts, where patients are unconscious or less able to actively participate in managing relationships, nurses acknowledged an increased need for nurse-family engagement. The goal of progressively engaging was to provide care that promotes patient and family well-being and enhances nurses' satisfaction in a job well done. 114 Many participants described nurses, patients, and family members managing their relationships so that they achieved a more personal level of engagement over time. Progressively engaging consisted of acts of coming together where nurses, patients, and family members moved forward on a trajectory of engagement; however, at any time nurses, patients, or family members could step back from active participation and move backward on the trajectory. Managing their relationships to try to achieve personal connections in the context of care in acute care hospital settings facilitated progressively engaging. Participants indicated that progressively engaging did not always occur. Nurses, patient, and family members maintained lower levels of engagement or stepped back from personal connections in response to systemic structural conditions, such as staffing shortages, or contextual structural conditions such as family not being present; personal factors such as attitudes and values held by nurses, patients, and family members that limited their desire to seek personal connections; and negative interpersonal dynamics occurring within the relationship. Nurses', patients' and family members' positive responses to structural conditions, personal factors, and interpersonal dynamics facilitated their efforts to achieve higher levels of engagement. 5.3 Stages of Engaging The process of progressively engaging includes stages of nurse, patient, and family interaction that can be characterized as progressively higher levels of engagement. The engagement trajectory reflects a lower level of personal connection at one end through to a higher level of personal connection at the other. Nurses, patients, and family members managed relationships by moving up or down the trajectory. If they moved down the trajectory by stepping back, there was the potential for families and patients to fall through the cracks or go 115 unnoticed except for receiving the most basic care in the form of task completion (see figure 1.0). The level of engagement attained when managing relationships depended on how nurses, patients, and family members individually and collectively perceived structural conditions and personal factors at any stage in the process. If nurses, patients, and family members perceived Figure 1.0. Progressively Engaging Process. Legend: N- nurse; F - family; P - patient 116 structural conditions and personal factors as favorable to managing relationships, they were more likely to pursue more personal connections, which moved their relationship to higher levels of engagement. As nurses, patients, and family members progressed along the trajectory toward higher levels of engagement their relationships became warmer and more reciprocal (Figure 1.0). Alternatively, when nurses, patients and family members perceived structural conditions and personal factors as unfavorable to managing their relationships they did not pursue more personal connections. They could not only resist moving to higher levels of engagement, but could also step back from active participation and commitment to each other and move to a lower stage of engagement. In those cases their relationships were characterized by impersonal, business-like and sometimes discordant interactions. The three stages of engagement in the progressively engaging process include: focusing on tasks, getting acquainted, and establishing rapport. These stages capture nurses', patients' and family member's efforts to manage their relationships. Each stage of managing relationships is depicted in Figure 2.0. In the section that follows, I provide a brief overview of the stages of the process of progressively engaging. 5.3.1 Stage one: Focusing on tasks. Focusing on tasks refers to the stage of minimal engagement. During this stage, nurses completed tasks that had to be done for the patient, such as administering medications, monitoring equipment, attending to basic physical needs of patients (e.g. feeding and bathing), and completing necessary documentation without acknowledging the recipients of care. Patients and family members focused on the competence with which nurses undertook the skills necessary to care for them and/or their ill family member. Interaction between nurses and 117 patients and family members was minimal and guarded. Any conversation tended to be patient-focused. At this stage, negative structural conditions and personal factors influenced the nurses', patients', and family members' management of their relationships by reducing the likelihood that they would move beyond focusing on tasks as a basis for interactions. If the terms of engagement were negative and were, in turn, followed by negative interpersonal dynamics, managing relationships at this stage was difficult and participants could step back from and/or opt out of the relationship altogether. Negative structural conditions included, for example, staff shortages. Personal factors pertained to a general lack of personal motivation on the part of the participants to pursue more personal connections for reasons such as previous bad experiences and personality traits. If, while completing nursing tasks, there were positive conditions that permitted the investment of time, and personal factors such as motivation, nurses, patients, and family members were more likely to move beyond a task focus as a basis for interaction to feel each other out. If while 'getting a feel for each other' the exchange of more personal information resulted in positive interpersonal dynamics, the group was more likely to move forward to the next stage of getting acquainted. If there were negative interpersonal dynamics, the group stayed at the impersonal or business-like level of interaction or "just doing the job". Moreover, if conditions and personal factors led to negative interpersonal dynamics, nurses, patients and family members could disengage even further and patients and family members "fell through the cracks". Falling through the cracks described managing relationships in ways where there were no attempts by the nurse to engage or interact beyond attending to basic physical care and medical monitoring needs. e r a e » Is) Nurse, patient, and family relationships: Just doing the job < *+ve 'terms'of • Stage 1: Focusing on Tasks [^•-engagement. •^ engagement -ve Interpersonal dynamics I +ve Interpersonal dynamics -ve Interpersonal dynamics Fall Feeling Fall through each through the other the cracks out cracks I I -ve +ve Interpersonal Interpersonal dynamics dynamics Progressively Engaging Stage 2: Getting Acquainted .• • i ve +ve Interpersonal Interpersonal dynamics dynamics i H i i i i i i i 1 * -1 Doing Stage 3: Establishing <' Rapport | IVThe VJob With I Heart IF I ' -ve Interpersonal dynamics +ve Interpersonal < dynamics 1 . i i i e r a a «> •t o e r a 2 •< n !~ W s e r a 88 e r a S' e r a Step i Back Step Back Step Back 119 When the way in which nurses, patients, and family members managed their relationships was characterized as 'just doing the job', with neutral or negative interpersonal dynamics, they described feeling dissatisfied because they felt disrespected, unappreciated, and undervalued. When getting an initial feel for each other was characterized by positive interpersonal dynamics they managed their relationships by investing more time and energy in order to get acquainted. In other words, initial interactions during focusing on tasks which demonstrated interest and concern for each other in relationship increased the likelihood that nurses, patients, and family members would move to a higher stage of engagement. 5.3.2 Stage two: Getting acquainted. When interest and concern for other(s) characterized interpersonal dynamics, the nurses, patients, and family members managed relationships by moving in