UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Constructing the experiences of people with advanced dementia who are nearing the end of life Puurveen, Gloria 2016

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

Item Metadata

Download

Media
24-ubc_2016_september_puurveen_gloria.pdf [ 4.94MB ]
Metadata
JSON: 24-1.0303147.json
JSON-LD: 24-1.0303147-ld.json
RDF/XML (Pretty): 24-1.0303147-rdf.xml
RDF/JSON: 24-1.0303147-rdf.json
Turtle: 24-1.0303147-turtle.txt
N-Triples: 24-1.0303147-rdf-ntriples.txt
Original Record: 24-1.0303147-source.json
Full Text
24-1.0303147-fulltext.txt
Citation
24-1.0303147.ris

Full Text

    CONSTRUCTING THE EXPERIENCES OF PEOPLE WITH ADVANCED DEMENTIA WHO ARE NEARING THE END OF LIFE  by  Gloria Puurveen  B.A., Kings University, 1991 B.M.T., B.C Open University, 1995 M.A., Simon Fraser University, 2009    A DISSERTATION SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES  (Interdisciplinary Studies) [Nursing/Audiology and Speech Language Sciences/Sociology]  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   May 2016   © Gloria Puurveen, 2016   ii    Abstract The purpose of this ethnographic case study research was to examine the experiences of people with advanced dementia nearing the end of life, how carers and family constructed and responded to the psychosocial, spiritual, and physical needs of the residents, and how discursive and contextual factors shaped residents’ experiences. The project comprised case studies of two women with advanced dementia who were identified as nearing the end of life, their carers, and family members. Over an 18-month period, I conducted over 600 hours of participant observation, 30 hours of video observation, and in-depth interviews with 23 carers and 2 family members in a nursing home in British Columbia, Canada. Informed by a dialogical perspective, I employed a case-oriented analysis to construct a rich textual and visual portrait of each woman’s life.  The first case study illustrated how carers and family understood the resident’s experiences according to four storylines: 1) the deteriorating body; 2) the institutionalized body; 3) the loved body; and 4) the ambiguous body. These storylines culminated in several tensions that underscore the challenges in caring for someone with advanced dementia nearing the end of life, with the notion of liminality providing a framework for discussing this case. The second case study focused on the co-construction of dialogical relationships between the resident, her carers and family. The findings demonstrated that interactions were marked by an ethical obligation towards the resident that went beyond the boundaries of fulfilling instrumental tasks to encountering and acknowledging the resident as a unique person. Findings are discussed in terms of an aesthetic approach to dementia end-of-life care. iii     The theoretical considerations of silence, the cultivation of knowledge, and the moral obligation for carers to nurture dialogical I-Thou relationships provided a framework for integrating the findings. Together, these case studies inform our understanding of the lived experiences of residents with advanced dementia who are nearing the end of life, and how carers and family come to understand and respond to these experiences.   iv    Preface This dissertation is an original intellectual product of the author G. Puurveen. The fieldwork reported in Chapters 4-8 was covered by UBC Behavioural Ethics Review Board ethics certificate H13-01913 for the project entitled: The end-stage of life: Experiences of persons with advanced dementia, their family, and the staff who care for them. Portions of Chapter Four in the sections Video Observations and Video Analysis and in Chapter Eight in the section Implications are modified from material published as a book chapter [Puurveen, G., Phinney, A., Cox, S.M., & Purves, B. (In Press). Ethical considerations in the use of video observations in dementia end-of-life care research. In D. Warr, M. Guillemin, S. Cox, S. and J. Waycott (Eds.), Ethics and visual research methods: Theory, methodology and practice. London: Palgrave Macmillan] and material published in the journal Visual Methodologies [Puurveen, G., Phinney, A., Cox, S.M., & Purves, B. (2015). Ethical issues in the use of video observations with people with advanced dementia and their caregivers in nursing home environments. Visual Methodologies, 3(2), 16-26]. I was responsible for concept formation and the majority of manuscript composition. A. Phinney, S.M. Cox and B. Purves contributed to manuscript edits.  All images and video included in this dissertation are used with the consent of study participants.   v    Table of Contents  Abstract	  ......................................................................................................................	  ii	  Preface	  .......................................................................................................................	  iv	  Table	  of	  Contents	  ........................................................................................................	  v	  List	  of	  Tables	  ...............................................................................................................	  x	  List	  of	  Figures	  .............................................................................................................	  xi	  List	  of	  Videos*	  ............................................................................................................	  xii	  List	  of	  Abbreviations	  ..................................................................................................	  xiii	  Acknowledgements	  ...................................................................................................	  xiv	  Dedication	  .................................................................................................................	  xvi	  Chapter	  One:	  Introduction	  ...........................................................................................	  1	  1.1	   Prelude	  ...................................................................................................................	  2	  1.2	   Background	  .............................................................................................................	  5	  1.3	   Research	  Problem	  ...................................................................................................	  8	  1.4	   Structure	  of	  Thesis	  .................................................................................................	  10	  Chapter	  Two:	  Review	  of	  Selected	  Literature	  ..............................................................	  12	  2.1	   Constructing	  Advanced	  Dementia	  Nearing	  the	  End	  of	  Life	  ......................................	  13	  2.1.1	   Defining	  Dementia	  ...................................................................................................	  13	  2.1.2	   Defining	  Nearing	  the	  End	  of	  Life	  ..............................................................................	  16	  2.1.3	   What	  is	  End-­‐of-­‐Life	  Care?	  ........................................................................................	  19	  2.2	   The	  Experience	  of	  Physical	  Symptoms	  ....................................................................	  21	  2.2.1	   Symptom	  Burden	  at	  the	  End	  of	  Life	  .........................................................................	  21	  2.2.2	   Responding	  to	  Symptom	  Burden	  .............................................................................	  24	  2.2.3	   Experience	  of	  Pain,	  Expressions	  of	  Pain,	  and	  Management	  of	  Pain	  ........................	  25	  2.2.4	   Dysphagia,	  Eating,	  and	  Mealtime	  Experiences	  ........................................................	  30	  2.3	   Social	  and	  Emotional	  Needs	  ...................................................................................	  32	  2.4	   Spiritual	  Needs	  .......................................................................................................	  35	  2.5	   Advance	  Directives	  .................................................................................................	  39	  2.6	   Advance	  Care	  Planning	  and	  Family	  Roles	  in	  Decision-­‐Making	  .................................	  41	  2.7	  	   Conclusion	  and	  Purpose	  of	  the	  Research	  ................................................................	  43	  Chapter	  Three:	  Theoretical	  Grounding	  .......................................................................	  46	  3.1	   A	  Persistent	  Perspective	  .........................................................................................	  46	  3.2	   A	  Counter-­‐Story	  .....................................................................................................	  48	  3.3	   A	  Dialogical	  Perspective	  .........................................................................................	  50	  3.3.1	   Co-­‐Authorship:	  The	  I,	  the	  Thou,	  and	  the	  Space	  Between	  ........................................	  51	  3.3.2	   Key	  Constructs	  of	  a	  Dialogical	  Perspective	  ..............................................................	  54	  3.3.2.1	   Utterances,	  Multi-­‐vocality	  and	  Dialogical	  Tensions	  ....................................................	  54	  vi    3.3.2.2	   Embodiment	  and	  Social	  Embodiment	  .........................................................................	  57	  3.3.2.3	   Emplacement	  and	  Social	  Situatedness	  .......................................................................	  63	  Chapter	  Four:	  Path	  of	  Inquiry	  ....................................................................................	  66	  4.1	   Introduction	  ...........................................................................................................	  66	  4.2	   A	  Constructivist	  Grounding	  ....................................................................................	  66	  4.3	   Strategy	  of	  Inquiry:	  Ethnographic	  Case	  Studies	  ......................................................	  67	  4.4	   Casing	  the	  Case	  ......................................................................................................	  71	  4.5	   Entry	  into	  the	  field	  .................................................................................................	  74	  4.6	   Recruitment	  Processes	  and	  Sample	  ........................................................................	  75	  4.6.1	   Recruiting	  Resident	  Participants	  ..............................................................................	  75	  4.6.2	   Recruiting	  Family	  Participants	  .................................................................................	  77	  4.6.3	   Recruiting	  Staff	  Participants	  ....................................................................................	  78	  4.7	   Data	  Collection	  Procedures	  ....................................................................................	  81	  4.7.1	   Participant	  Observation	  ...........................................................................................	  81	  4.7.1.1	   Observational	  Field	  Notes	  as	  a	  Source	  of	  Data	  ...........................................................	  84	  4.7.1.2	   Casual	  Conversation	  as	  a	  Source	  of	  Data	  ....................................................................	  85	  4.7.2	   Interviews	  ................................................................................................................	  86	  4.7.2.1	   Participant	  Interviews	  .................................................................................................	  88	  4.7.2.2	   Structure	  of	  the	  Interviews	  .........................................................................................	  88	  4.7.3	   Video	  Observation	  ...................................................................................................	  92	  4.7.3.1	   Video	  Footage	  as	  a	  Source	  of	  Data	  .............................................................................	  94	  4.7.3.2	   Equipment	  ...................................................................................................................	  95	  4.7.3.3	   Video	  Reflexivity	  as	  a	  Source	  of	  Data	  ..........................................................................	  96	  4.7.4	   Document	  Review	  ....................................................................................................	  99	  4.8	   Analytical	  Procedures	  ............................................................................................	  99	  4.8.1	   Case-­‐Oriented	  Analysis	  ............................................................................................	  99	  4.8.2	   Process	  of	  Analysis	  .................................................................................................	  103	  4.8.3	   Analysis	  of	  Textual	  Data:	  Fieldnotes	  and	  Casual	  Conversations	  ............................	  103	  4.8.4	   Video	  Analysis	  ........................................................................................................	  105	  4.8.4.1	   The	  Video-­‐Editing	  Process	  .........................................................................................	  107	  4.8.5	   Analysis	  of	  Interview	  Recordings	  ...........................................................................	  109	  4.8.6	   Document	  Analysis	  ................................................................................................	  109	  4.9	   Reflexivity	  and	  Ethical	  Research	  Practice	  ..............................................................	  110	  4.9.1	   Rigour	  and	  Credibility	  ............................................................................................	  111	  4.9.2	   Meaningful	  Coherence	  ..........................................................................................	  112	  4.9.3	   Sincerity	  Cultivated	  Through	  Reflexive	  Research	  Practice	  ....................................	  112	  4.9.3.1	   Representation	  and	  Voice	  .........................................................................................	  115	  4.9.4	   Seeking	  Rigour	  through	  Ethical	  Practice	  ................................................................	  117	  4.9.4.1	   Consent	  Process	  ........................................................................................................	  118	  4.9.4.2	   Resident	  Assent	  .........................................................................................................	  119	  4.9.4.3	   Consent	  and	  Assent	  for	  Staff	  Participation	  ...............................................................	  121	  4.9.4.4	   Confidentiality	  and	  Privacy	  .......................................................................................	  122	  4.9.5	   Emergent	  Ethics	  and	  In-­‐situ	  Research	  Practice	  .....................................................	  123	  4.9.5.1	   Negotiating	  Roles	  and	  Relationship:	  .........................................................................	  124	  4.9.5.2	   Researcher’s	  Influence	  on	  the	  Performance	  of	  Care:	  ...............................................	  126	  4.9.5.3	   The	  Emotional	  Toll	  of	  Researching	  Sensitive	  Topics	  .................................................	  128	  Chapter	  Five:	  Anchoring	  Points	  ................................................................................	  129	  5.1.	   Introduction	  .........................................................................................................	  129	  vii    5.2	   	  The	  Physical	  and	  Organizational	  Environment	  .....................................................	  129	  5.2.1	   The	  Main	  Floor	  .......................................................................................................	  130	  5.2.2	   The	  Complex	  Care	  Units	  ........................................................................................	  131	  5.2.3	   The	  Special	  Care	  Unit	  .............................................................................................	  133	  5.3	   Social	  and	  Spiritual	  Environment:	  The	  Making	  of	  Community	  ..............................	  135	  5.4	   Summary	  .............................................................................................................	  139	  Chapter	  Six:	  Ella	  ......................................................................................................	  140	  6.1	   Introduction	  .........................................................................................................	  140	  6.2	   Narrative	  Beginnings	  ............................................................................................	  142	  6.3	  	   The	  Deteriorating	  Body	  ........................................................................................	  144	  6.3.1	   The	  Physical	  Reality	  of	  a	  Deteriorating	  Body	  .........................................................	  144	  6.3.1.1	   Skin	  and	  Bones	  ..........................................................................................................	  144	  6.3.1.2	   Vision	  and	  Hearing	  ....................................................................................................	  146	  6.3.1.3	   Eating	  and	  Swallowing	  ..............................................................................................	  147	  6.3.1.4	   Bowels	  .......................................................................................................................	  148	  6.3.2	  A	  Mechanically-­‐Mediated	  Existence:	  Wheelchairs	  and	  the	  Deteriorating	  Body	  .......	  148	  6.3.3	   Summary	  ................................................................................................................	  151	  6.4	   The	  Institutionalized	  Body	  ...................................................................................	  152	  6.4.1	   Structuring	  the	  Everyday	  Reality:	  Expectations,	  Control	  and	  Routine	  ..................	  152	  6.4.2	   The	  Institutionalized	  Body	  and	  the	  Discursive	  Environment	  .................................	  155	  6.4.2.1	   Discipline,	  Labels,	  and	  Body	  Parts	  .............................................................................	  155	  6.4.2.2	   The	  Institutionalized	  Body	  as	  Reflected	  in	  Verbal	  Utterances	  ..................................	  156	  6.4.3	   Just	  a	  Body	  in	  Care:	  The	  Objectified	  and	  Overlooked	  Body	  ...................................	  157	  6.4.4	   Resisting	  Being	  an	  Institutionalized	  Body	  ..............................................................	  161	  6.4.5	  	   Summary	  ...........................................................................................................	  163	  6.5	   The	  Loved	  Body:	  Bodies	  in	  Relationship	  ...............................................................	  164	  6.5.1	   The	  Storied	  Body	  ...................................................................................................	  164	  6.5.2	   The	  Witnessed	  Body	  ..............................................................................................	  168	  6.5.2.1	   Touch	  and	  Presence	  ..................................................................................................	  168	  6.5.2.2	   Creating	  Meaningful	  Moments	  .................................................................................	  170	  6.5.2.3	   Attention	  to	  the	  Aesthetic	  Body	  ...............................................................................	  171	  6.5.3	  	   The	  Known	  Body:	  Bodies	  in	  Care	  .......................................................................	  172	  6.5.3.1	   The	  Acknowledged	  and	  Respected	  Body	  ..................................................................	  172	  6.5.3.2	   Relational	  Knowing	  ...................................................................................................	  173	  6.5.3.3	   A	  Feeling	  for	  the	  Work:	  Sensory	  Knowing	  .................................................................	  175	  6.5.4	   Summary	  ................................................................................................................	  177	  6.6	   Interlude:	  Ave	  Maria	  ............................................................................................	  178	  6.7	   	  The	  Ambiguous	  Body	  ..........................................................................................	  178	  6.7.1	   The	  Communicating	  Body	  ......................................................................................	  179	  6.7.1.1	   On	  Talking	  .................................................................................................................	  179	  6.7.1.2	   On	  Crying	  ..................................................................................................................	  180	  6.7.2	   Interpreting	  Vocal	  and	  Bodily	  Expressions	  ............................................................	  182	  6.7.2.1	   Pain,	  Distress,	  and	  the	  DS	  Storyline	  ..........................................................................	  183	  6.7.2.2	   Food	  Refusal	  versus	  Food	  Acceptance	  ......................................................................	  186	  6.7.2.3	   Social	  Withdrawal:	  Becoming	  Silent,	  Becoming	  Still	  .................................................	  188	  6.7.3	   Ambiguous	  Dying	  ...................................................................................................	  190	  6.7.4	   Summary	  ................................................................................................................	  193	  6.8	   Spaces	  of	  Tension	  .................................................................................................	  194	  6.8.1	   Keeping	  Death	  at	  Bay:	  Eating,	  Feeding-­‐Assists,	  and	  Agency	  .................................	  194	  viii    6.8.2	   Bearing	  Witness	  .....................................................................................................	  197	  6.8.2.1	   The	  Solitude	  of	  Suffering:	  Witnessing	  Ella’s	  Crying	  ...................................................	  197	  6.8.2.2	   Facilitating	  Comfort:	  Moral	  Distress	  and	  the	  Institutionalized	  Body	  ........................	  200	  6.8.3	   Seeking	  Balance	  .....................................................................................................	  201	  6.9	   Summary	  .............................................................................................................	  205	  6.10	   Commentary:	  Liminality	  and	  Watchful	  Waiting	  ....................................................	  207	  Chapter	  Seven:	  Elizabeth	  .........................................................................................	  214	  7.1	   Introduction	  .........................................................................................................	  214	  7.2	  	   Narrative	  Beginnings	  ............................................................................................	  216	  7.2.1	   The	  Move-­‐in	  Story:	  (Un)settling-­‐in	  ........................................................................	  216	  7.2.2	   A	  Deteriorating	  Body	  .............................................................................................	  219	  7.2.3	   New	  Beginnings:	  The	  Beginnings	  of	  the	  Social	  Body	  .............................................	  220	  7.2.4	   Summary	  ................................................................................................................	  221	  7.3.	   Co-­‐Constructing	  Embodied	  Selfhood	  ....................................................................	  222	  7.3.1	   Introduction	  ...........................................................................................................	  222	  7.3.2	   Gestural	  Repertoire	  ...............................................................................................	  223	  7.3.3	   Inward-­‐Directed	  Expressions	  .................................................................................	  224	  7.3.3.1	   Description	  of	  Inward-­‐Directed	  Expressions	  .............................................................	  224	  7.3.3.2	   Interpreting	  Inward-­‐Directed	  Expressions	  ................................................................	  225	  7.3.4	   Outward-­‐Directed	  Expressions	  ..............................................................................	  226	  7.3.4.1	  Description	  of	  Outward-­‐Directed	  Expressions	  .................................................................	  226	  7.3.4.2	   Interpreting	  Outward-­‐Directed	  Expressions	  .............................................................	  229	  7.3.4	   Kinesthetic	  Reflexivity	  ...........................................................................................	  230	  7.3.4	   Summary	  ................................................................................................................	  231	  7.4	   Typical	  Interactions	  During	  a	  Typical	  Day	  .............................................................	  232	  7.5	   Being	  Loved:	  Interactions	  with	  Thomas	  ................................................................	  233	  7.5.1	   Maintaining	  Patterns	  of	  Connection	  ......................................................................	  234	  7.5.2	   An	  Acknowledged	  Relationship	  .............................................................................	  236	  7.5.3	   Being	  in	  Relationship	  .............................................................................................	  238	  7.6	   Being	  Acknowledged:	  Fleeting	  Interactions	  ..........................................................	  239	  7.6.1	   Not	  Dependent	  on	  Outward-­‐directed	  Expressions	  ................................................	  241	  7.6.2	   Given	  Positive	  Regard	  ............................................................................................	  242	  7.7	   Being	  Nurtured:	  On	  Rituals	  and	  Rock	  n’	  Roll	  .........................................................	  243	  7.7.1	   Honky-­‐Tonk	  Woman	  ..............................................................................................	  245	  7.7.2	   Peace	  Be	  with	  You	  .................................................................................................	  246	  7.7.3	   Being	  Part	  of	  a	  Community	  ....................................................................................	  249	  7.7.4	   Being	  Part	  of	  an	  Aesthetic	  Environment	  ................................................................	  249	  7.7.5	   Being	  Nurtured	  ......................................................................................................	  251	  7.8	   Being	  Nourished:	  Mealtime	  Interactions	  ..............................................................	  253	  7.8.1	   Nourishment:	  A	  Focus	  on	  Task	  ..............................................................................	  253	  7.8.2	   Nourishment:	  A	  Focus	  on	  Relationship	  .................................................................	  255	  7.9	   Being	  Cared	  For:	  Personal	  Care	  Interactions	  .........................................................	  258	  7.9.1	   The	  Structure	  of	  Personal	  Care	  Encounters	  ...........................................................	  259	  7.9.2	   Appearance	  Work	  ..................................................................................................	  261	  7.9.3	   The	  Nature	  of	  Verbal	  Exchanges	  ............................................................................	  263	  7.9.4	   General	  Characteristics	  of	  Nonverbal	  Gestures	  .....................................................	  268	  7.9.4.1	   The	  Energy	  Between:	  Reading	  the	  Body	  and	  Nonverbal	  Interactions	  ......................	  271	  7.9.4.2	   The	  Art	  of	  Care	  Coordination	  ....................................................................................	  275	  ix    7.10	   	  Summary	  ............................................................................................................	  280	  7.11	   Commentary:	  The	  Aesthetic	  of	  Care	  .....................................................................	  281	  Chapter	  Eight:	  Discussion	  ........................................................................................	  284	  8.1	   Co-­‐Constructing	  I-­‐Thou	  Relationships	  ...................................................................	  285	  8.1.1	   From	  Talk	  to	  Silence:	  Embodied	  Communication	  in	  I-­‐Thou	  Relationships	  ............	  285	  8.1.2	   Paths	  of	  Knowing	  ...................................................................................................	  290	  8.1.2.1	   Sensory	  Knowing	  .......................................................................................................	  292	  8.1.2.2	   Emplaced	  Knowing	  ....................................................................................................	  294	  8.1.2.3	   Relational	  Knowing	  ...................................................................................................	  297	  8.1.2.4	   Narrative	  Knowing	  ....................................................................................................	  299	  8.1.3	   Moral	  Tasks	  and	  Moral	  Relationships:	  A	  Call	  to	  the	  Face	  of	  the	  Other	  .................	  301	  8.1.3.1	   Revisiting	  I-­‐It	  and	  I-­‐Thou	  Modes	  of	  Relating	  .............................................................	  302	  8.1.3.2	   Emplaced	  Relationships	  ............................................................................................	  304	  8.1.3.3	   Reciprocity	  and	  the	  “Work”	  of	  Asymmetry	  ...............................................................	  306	  8.2	   Study	  Implications	  ...............................................................................................	  312	  8.2.1	   Embodied	  Selfhood	  at	  the	  End	  of	  Life	  ...................................................................	  312	  8.2.2	   Accentuating	  Paths	  of	  Knowing	  and	  Embodied	  Communication	  ..........................	  314	  8.2.3	   Ethical	  Issues	  in	  the	  use	  of	  Video	  Observations	  ....................................................	  317	  8.3	   Study	  Limitations	  .................................................................................................	  319	  8.4	   Future	  Research	  ...................................................................................................	  322	  8.5	   Conclusion	  ...........................................................................................................	  323	  References	  ..............................................................................................................	  325	  Appendix	  A:	  Sample	  Recruitment	  Letter	  ..................................................................	  370	  Appendix	  B:	  Sample	  Consent	  Form	  ..........................................................................	  372	  Appendix	  C:	  Interview	  Guide	  ...................................................................................	  379	  Appendix	  D:	  Field	  Notes	  ..........................................................................................	  381	    x    List of Tables Table 1: Carer Sample Characteristics ...................................................................79	  Table 2: Verbal Utterances in a Personal Care Encounter ...................................264	   xi    List of Figures Figure 1: Her Eyes Light Up ................................................................................214	  Figure 2: Gestures and Expressions .....................................................................223	  Figure 3: Being Acknowledged ...........................................................................239	  Figure 4: Being Nurtured .....................................................................................248	  Figure 5: Being Nourished ...................................................................................255	  Figure 6: Appearance Work .................................................................................262	  Figure 7: Being Cared For ...................................................................................270	  Figure 8: Affixing the Splints ..............................................................................274	     xii    List of Videos* Video 1: Prelude ..................................................................................................140	  Video 2: Interlude: Ave Maria .............................................................................178	  Video 3: Postlude: Breath ....................................................................................213	                   * Video clips have restricted distribution and are not available as an embedded moving image in the electronic version of this dissertation.  xiii    List of Abbreviations ADLs Activities of daily living AH Allied health staff BAN-S Bedford Alzheimer Nursing Scale-Severity Subscale  BPSD Behavioural and Psychological Symptoms of Dementia CA Care Aide CCU Complex Care Unit CIHI Canadian Institute of Health Information CPS Cognitive Performance Scale DOI Degree of intervention DNH Do-not-hospitalize FAST Functional Assessment Staging Tool GDS Global Deterioration Scale LPN Licensed Practical Nurse MSSE Mini Suffering Status Exam MT Music Therapist NS Nursing Staff PRN Pro re nata (when necessary) RAI-MDS 2.0 Resident Assessment Instrument – Minimum Data Set RN Registered Nurse RT Recreation Therapist SCU Special care unit WHO World Health Organization  xiv    Acknowledgements First and foremost, I thank the residents, their families, and the carers who participated in this research. Words are inadequate to express the depth of my gratitude to these individuals for allowing me to be part of their lives. I thank them for their trust, their presence, and their honesty. They have touched me deeply, and I thank them for all they have done.  I am immensely appreciative of the support and guidance from my supervisory committee. Dr. Alison Phinney (PhD co-supervisor), I thank you for your steadfastness, your patience and faith in me, and your infinite wisdom and generosity of spirit. Dr. Barbara Purves (PhD co-supervisor), I thank you for your intellectual rigour, your wisdom, your critique and edits of my numerous drafts, and your use of Tolkien metaphors. Dr. Susan Cox, I thank you for engaging with me intellectually, for nurturing my creative voice, and for your ongoing encouragement and support. I thank you all for continually challenging my thinking and contributing to my development as a scholar. I acknowledge the University of British Columbia whose funding support allowed me to conduct this research.  I am fortunate to have a tremendous network of friends who have supported and encouraged me throughout this process. I give thanks to Heather whose insights were worth its weight in gold, and to Bindy whose intellectual passion inspired my own. My family has been from the beginning a source of love, joy and acceptance. I thank them for their abiding support, their listening ear, and for sustaining me in so many ways.  xv    Finally, I express my deepest gratitude towards Robert Gill, whose love and unwavering belief in me allowed me the space to do this work. I could not have done this without you.  xvi    Dedication  To Rob, my beloved.1    Chapter One: Introduction               You would know the secret of death. But how shall you find it unless you seek it in the heart of life? The owl whose night-bound eyes are blind unto the day cannot unveil the mystery of light If you would indeed behold the spirit of death, open your heart wide unto the body of life. For life and death are one, even as the river and the sea are one  (Gibran, 1923, p.90)  2    1.1 Prelude “I don’t think she’s going to die tonight,” Julia says, “why don’t you go home and come back in the morning.” It was just after 11 pm and having been at the care facility all day, I was tired but hesitant to leave. Molly had rallied somewhat. She seemed more vital than she had all day—her eyes were open, she moved about in bed and was opening her mouth for a drink of water. She seemed more relaxed than she had during the day and her fever was down.   But things change, as they so often do. Molly died at 7:45 the following morning. Alone. Staff were in morning report. I arrived 15 minutes later.   Clare, a casual LPN, is incharge. I find her at the nursing station searching for the number of the funeral home. “What happened?” I ask.   “I went to check on Molly, and noticed that she wasn’t breathing. I checked her pulse and there was none.  I then went upstairs to get the RN to pronounce her dead.” As she speaks, Clare clutches at her chest and her face flushes.  “How was that for you? Finding her like that?” I ask.  Clare shrugs, “I’m not unfamiliar with death…it doesn’t bother me too much.” She takes a deep breath and clutches her fist even tighter. A lingering pause, “you need to protect yourself otherwise…you can’t cry and cry and cry or you….your har…” Her words falter.  “You’re heart starts to break?” I suggest.   With eyes tearing, Clare says, “Yes, your heart starts to hurt and you can’t care for all the other residents who need you in that moment. But, this one, and this one—they need their breakfast and that one needs to be helped out of bed.” She takes a deep breath and unclutches her hand and smiles.   Indeed, life in the care facility goes on. Breakfast is about to be served.  ~~~~~~  Two weeks prior to Molly’s death   I was in the recruitment phase of my study when I met with Molly’s son, her proxy decision-maker, to recruit her for my study. I explained to him that I was interested in understanding Molly’s everyday experiences, how carers constructed her experiences, and how care responded and shifted as her physical, psychosocial, and spiritual needs changed.  I did not expect change to happen so quickly. 3     ~~~~~~~~~~  One week prior to Molly’s death  I meet with the care aides. “You’re going to observe Molly?” Alice, a care aide asks.  “Yes,” I reply.  Alice chuckles and tells me that Molly was quite ill about 7 months ago, but she rallied back and appears to be doing really well. I ask the care aides the “would you be surprised” question. The room is divided: some think Molly will die within the next six to 12 months while others think longer. I wonder if I had been too hasty to recruit Molly.  The following day Molly is sent to the hospital, despite having a do-not-hospitalize directive. She has an internal bleed, likely caused by cancer—though this is not an official diagnosis. She is admitted overnight but no further investigation is conducted as per the family’s wishes. Molly returns to the facility, her care plan updated to reflect her change in status to palliative. It is anticipated that Molly has a few more days to live.   ~~~~~~  The evening prior to Molly’s death  Coming into the care home at noon, I am greeted by Julia, a nurse, who tells me that Molly has a fever, and is no longer taking anything by mouth. Steeling myself for the inevitable, I walk into Molly’s room. She is in bed, dressed in a blue hospital gown and covered with a red quilt. She looks different—her face sallow and sunken, her head shrunken into her shoulder. Her breath is audible, shallow, fast and crackling. The spectre of death seems to be lurking in the corners. I shiver. The curtains that dress the large picture windows are open, revealing a brilliant blue sky. Crows hang out on the ledge of the building—their faces turned towards the room, looking watchful. Music wafts into the room from the stereo in the lounge down the hall. Choral music.  It sounds Baroque.  Maya, a care aide, walks into the room, shaking her head.  “I can’t believe it,” she says, “She ate her whole meal last night.” Earlier in the week, Maya had alerted me to Molly pushing out the food with her tongue. “This is a sign that she doesn’t want it,” she says to me.  A while later, Alice walks into the room. “How is she doing?” she asks? I shrug. I honestly don’t know what to say. How does one answer the question in the context of someone who is dying?  “This happened not so long ago, and she pulled through.” She smiles at Molly, who does not seem aware of her presence. But Alice looks doubtful, as if she failed to convince herself with the idea that Molly might rally back.  Extra care has been arranged. A care aide will sit with Molly over the evening and into the night and keep vigil. The extra care aide, Stella, has arrived and is hesitant to be in the space. As Julia inserts two subcutaneous lines into Molly’s arms, one for morphine and 4    one for glycopyrolate, Stella, gesturing to the activity, tells me that this is very different than what she is familiar with in her home country. She explains that when a person needs palliative care, they are discharged home and the family takes up the care.    This might be the reason why she seems so perplexed that her role is “added care.” She tells Julia that she’s not sure what she is supposed to do. Julia seems equally mystified by the question and rattles a litany of things to do: “you can massage her, hold her hand, talk to her, sing to her, if you want you can pray with her…you can massage lotion on her body as her skin is really dry, ah, anything to make her not feel alone.”  Stella decides to put lotion on Molly’s legs. As she does this, I think about all the activity that Molly did; how strong her legs once were. At age 55 she took up cycling and cycled across Canada, around Ireland and throughout the United States—a feat that most people cannot claim.   A while later Julie comes into the space to clean Molly’s fingernails, which are already clean and neatly manicured. She finishes her task and crouches beside the bed near Molly’s face, gently touching her face with the tips of her fingertips, caressing her forehead and neck and smoothing back Molly’s hair. No words are said. A stillness descends on the room transcending the sounds of a busy care facility environment. It’s only a moment, but so much seems to be said in that light caress.  All evening a steady stream of care staff enter and exit the space. Some pop their head in for a moment and others linger beside Molly’s bed, their gaze fixed intently on her, standing beside her in silent communion. Some share stories of their experiences of caring for Molly, and what is clearly evident is the love that they have for her.  “She is very special to us,” Julie comments, “She is part of my life.”   Stephen Post (2003) writes, Love is manifest in care, which is love in response to the other in need; it is manifest in compassion, which is love in response to the other in suffering; it is manifest in companionship, which is love attentively present with the other in ordinary moments (p. 30). Ordinary moments. I think back to Clare’s comment and how she is used to death.  She certainly would be.  She is a nurse in a residential care home, and in some way, the death of a resident is an ordinary moment in the life of residential care.   And yet, it is also an extraordinary moment. In the hours of witnessing Molly’s last moments of her life, I reflect that in this present moment, Molly is surrounded in love. A love generously given.  5    1.2 Background Dementia is a life-limiting illness that affects 35.6 million people around the world (World Health Organization [WHO], 2012). In Canada, by 2038, it is expected that 1.1 million Canadians will be living with dementia (Alzheimer Society of Canada, 2010). As the number of people living with dementia is expected to rise precipitously over the next few decades, so too will the demand for residential care increase (Alzheimer Society of Canada, 2010). A recent report from the Canadian Institute of Health Information (CIHI, 2014-2015) estimated that 61.2% of individuals living in long-term care facilities have Alzheimer’s disease or a related dementia.  In British Columbia (BC), 27.2% of all deaths occur in a residential care facility (CIHI, 2007); however, end-of-life care has been found wanting. Indeed, a recent BC ombudsperson’s report (2012) stated, “we expected residential care facilities to be models of excellence when it comes to planning for and providing end-of-life care. We did not find this to always be the case” (p. 305). Engle (1998) suggested that good end-of-life nursing home care involves a focus on daily routines, pain relief, attention to swallowing challenges, the availability of religious activities, and respectful and prompt care. Yet, a number of studies assessing the quality of end-of-life care in nursing home environments indicate a need for improvement (Oliver, Porock, & Zweig, 2005). Heavy staff workloads and high resident acuity (Cammer et al., 2014; Kayser-Jones et al., 2003), limited staff knowledge and resources (Goddard, Stewart, Thompson, & Hall, 2013; Kuhn & Forrest, 2012), an unsupportive physical environment (Kayser-Jones et al., 2003), and lack of attention to resident psychosocial and spiritual needs (Komaromy, Sidell, & Katz, 2000) contribute to the complexity of end-of-life care for all who live in residential care.  6    It is widely acknowledged that care for individuals with advanced dementia who are nearing the end of life is complex, challenging, and at times, contested. Yet, only recently has the trajectory of advanced dementia been articulated (Mitchell et al., 2009). From a biomedical perspective, excellent scholarship exists to inform advanced dementia end-of-life care. While translation of this research into health care practice is still in its infancy, research findings regarding pain management (Hendriks, Smalbrugge, Hertough, & van der Steen, 2014), the use of artificial nutrition and hydration (Gillick & Volandes, 2008), antibiotic use (Givens, Jones, Shaffer, Kiely, & Mitchell, 2010), advance directives (Mitchell, Morris, Park & Fries, 2004), and family satisfaction with care (Caron, Griffith & Arcand, 2005) provide an evidence base on which to provide quality end-of-life care. This research base however fails to explore the psychosocial and spiritual dimensions of dementia end-of-life care and as such, while there is a growing understanding of the physical/functional dimensions of the dying experience, there is much less of an understanding of other dimensions.  Moreover, access to quality end-of-life care for people with advanced dementia is limited (Small, Froggatt, & Downs, 2007) and research outlines several barriers to timely access to palliative care. First, with a slow trajectory of deterioration, prognosis in dementia remains elusive (Lynn & Adamson, 2003). While Molly received end-of-life care in the last week of her life, at the time of recruitment into this study, many staff thought that Molly would live more than a year.   Second, communication breakdown between family and care providers (Caron et al., 2005) and the lack of staff awareness to the innuendos of death (Parker & Macleod, 2001) lead to challenges with advance directives and end-of-life care planning. This was 7    observed in Molly’s case as well. Molly experienced a sudden change in health status (unexplained bleeding), changing the course of her care. She was hospitalized despite having a do-not-hospitalize (DNH) order, and subsequently returned to the facility with an order for palliation.  Third, in consideration of the care aide who kept vigil by Molly’s bedside, the notion of palliative care might be at odds with what a care staff may be familiar with, or what she believes. This points to a larger systems issue of the lack of understanding of palliative care and the benefits for people with advanced dementia (Sanders & Swails, 2009), and it underscores inadequate education of nursing home staff (Jones et al., 2004; Raymond, Warner, Davies, & Baishnab, 2013)  Yet, research also demonstrates that palliative care is associated with improved comfort for the person with advanced dementia (for example, better pain management; Hendriks et al., 2014) and more favourable family satisfaction with care (Caron et al., 2005). As reflected in the WHO’s (2002) definition below, palliative care considers the needs of the whole person and, by its very nature, is a person-centred approach to care (Small et al., 2007; van der Steen et al., 2014): an approach that improves the quality of life of patients and their families facing the problems associated with life threatening illness, through the prevention and relief of suffering by means of early identification and impeccable assessment and treatment of pain and other problems, physical, psychological and spiritual (p. 83). While access to quality palliative care is limited as noted above, what seems to be missing from the advanced dementia end-of-life literature is research oriented towards understanding psychosocial and spiritual dimensions of a person’s experience. 8    For example, the evening before Molly’s death, a steady stream of care staff visited with her, either sitting in silence or sharing stories. Molly was situated within a web of relationship fostered over the 12 years she lived at the facility. The care staff knew her story, her passion for cycling, and her sporting achievements. By weaving a narrative web from the threads of Molly’s life, care staff, in some way, facilitated closure and respectfully bore witness to the final moments of Molly’s life.   Likewise, psychosocial and spiritual dimensions also relate to the ways in which care staff understand and encounter death. Molly’s story brings to the fore the work that needs to be done in supporting staff who confront dying and death. While death is a typical event in the nursing home environment, it also is a profound event, and not acknowledging care staff’s experiences of death and dying undermines the relationships they have cultivated with the residents, and undermines their personhood (Kitwood, 1997). 1.3 Research Problem This research is motivated by a fairly straightforward question: what are the experiences of people with advanced dementia nearing the end of life and those who care for them? As discussed, current discourse is predominated by biomedicine that constructs a storyline of an experience dominated by the decline and decay of the body. Along the same storyline are the narratives related to the delivery of end-of-life care decision-making and family reflections of end-of-life care. These narratives are also primarily constructed in terms of biomedicine and instrumental responsibilities such as limiting futile medical intervention and care tasks that are oriented towards caring for the dying body.   9    While the dying body is a “corporeal reality…[that shapes] the dying experiences and actions of those charged for caring for the dying person (Seymour, Gott, Ahmedzai, & Clark, 2002, p. 289), the focus on the dying body neglects the psychosocial, emotional, and spiritual elements of the dying experience. If we accept that the dying experience is more than an experience of a deteriorating body (Kellehear, 2009), then the narrative of an embodied reality that highlights the spiritual, emotional, and relational experience is also important to consider. Moreover, if we accept that the person’s experience is influenced by the enactment of organizational procedures (e.g., decision-making), then the narrative of the ways in which caregivers, family and the institutional environment intersect and story that experience also needs to be addressed. That is, we also need to account for the multidimensional and dynamic interplay of the various discursive practices that influence the experience of dying from advanced dementia.  Reflecting on the shifts in the causes of death and the medical care thereof, Lynn (2005) writes: “Things have changed so much that today we don’t really have the language, the categories, and the stories to help us make sense of our situation” (p. S14). The goal of this research is to contribute to an emerging language that complements, but does not stand outside of, the dominant biomedicine discourse (i.e., the language of the dying body cannot be neglected). Moreover, it seeks to offer an intimate and deeply contextual narrative that helps make sense of dying from advanced dementia from the purview of multiple co-authors of the experience. 10    1.4 Structure of Thesis This thesis is the culmination of my involvement in the lives of two women1 with advanced dementia, their families, and the staff who cared for them. What is written arises out of observations of the two women and the dialogues between myself and family and care staff that highlights a shared understanding of advanced dementia, dying, and death. In the process of retracing and re-telling the events and stories, I am influenced by the commitment to knowing about and sharing these women’s experiences and elucidating what the care staff do, all with a deep recognition that what is portrayed is only a partial understanding of the story (Frank, 2005). This thesis is structured into eight chapters. The next chapter, Review of Selected Literature, begins with a discussion on defining advanced dementia nearing the end of life, which grounds the methodological decision made around the parameters of defining the case. I then move on to a discussion of relevant literature about end-of-life care in the context of advanced dementia and the experiences of the person with advanced dementia, their family, and the staff who care for them.  Chapter three, Theoretical Perspectives, provides the basis for foregrounding the concepts of embodiment, dialogical relationships, and emplacement, which are articulated through the theoretical lens of dialogical theory, situated body agents and embodied knowledge.  Chapter four, Path of Inquiry, begins with the epistemological orientation of the study, followed by a discussion of case study methods. I then go into detail about the                                                 1 For reasons discussed in section 4.6.1 Recruiting Resident Participants, Molly, the subject of the opening of this thesis, is not one of the subjects of the findings. 11    procedures of the study: entrée and recruitment, data considerations and analytical procedures. The chapter closes with a discussion about ethical issue that arose through the study. Chapter five, Anchoring Points, introduces the findings and provides a description of the place in which this study was conducted. In addition, it examines the socio-organizational context that is relevant to both cases. Although there is some overlap in the findings, chapters six and seven present each case separately. I was motivated to do this for two reasons: First, methodologically, it made more sense to maintain the integrity of each case as a conceptual whole, rather than moving back and forth between cases, which could cause confusion. Second, to honour both women’s stories demanded that their data be presented separately. Both chapters include visual images woven together with the written word. In the final chapter, I integrate findings from both cases and consider their significance in the context of relevant scholarly literature and theoretical orientations. The chapter concludes with a discussion of the limitations of the study, implications, and new questions arising from this research.    12    Chapter Two: Review of Selected Literature In this chapter, I provide the context for the study by addressing the research, issues and debates that circulate around advanced dementia nearing the end of life, nursing home dementia end-of-life care and family experiences. I first describe the construction of advanced dementia as a disease category and the construction of nearing the end of life. Next, I present the literature that addresses the experiences of people with advanced dementia nearing the end of life, particularly looking at the empirical research that addresses symptom burden, pain and swallowing problems. I then examine the literature related to socio-emotional and spiritual experiences. Finally, I discuss the use of advance directives and advance care planning. The literature accessed for this review had to be conceptualized as an advanced dementia end-of-life study or an end-of-life study that included those with advanced dementia as part of the sample study population. The search terms used included: advanced, severe or end-stage dementia and palliative, hospice or end-of-life care. Consequently, I did not examine the literature that explored experiences of those with advanced dementia outside the end-of-life context (see for example, Perrin, 1999), or research that examined general nursing home end-of-life care (see for example, Kayser-Jones, 2002; Kayser Jones et al., 2003). This exclusion will be especially notable in terms of psychosocial and spiritual care where there is a paucity of research framed within the end-of-life context.  13    2.1 Constructing Advanced Dementia Nearing the End of Life 2.1.1 Defining Dementia  Dementia, as a disease category, has been largely understood through the lens of biomedicine and is primarily described as a syndrome caused by progressive, irreversible neuro-degeneration characterized by cognitive and functional deficits. The disease manifests in symptoms of significant impairment of memory and executive functioning, visual/perceptual changes and language impairments, and impaired ability to do activities of daily living (ADLs). In addition to cognitive impairment, many individuals will develop personality changes and behavioural and psychological symptoms of dementia (BPSD) such as agitation, depression, psychotic symptoms, and aggression (WHO, 2012). Dementia takes many forms, the most common being Alzheimer disease, followed by vascular dementia, Lewy body disease, and fronto-temporal dementia as well as other less common subtypes (WHO, 2012). However, the forms of dementia, as they arise in the end-of-life literature, are typically undifferentiated (Hughes, 2011).  The experiences of people with advanced dementia are constructed along storylines of cognitive and functional loss and the potential for distressing physical symptoms. The person will have lost verbal capabilities, though she may still have the capacity to vocalize or emit single words. She will have lost her ability to complete ADLs such as bathing, toileting, dressing, and personal hygiene, and thus is completely dependent on others for support. She will have lost her ability to walk, hold her body upright and adjust her postural position. Leading up to her death, pain, constipation, pressure ulcers, and infections are common (Mitchell et al., 2009), as are behavioural challenges and suffering (Aminoff, 2012). 14    In spite of the presentation of similar characteristics that portray advanced dementia, the term advanced dementia is inconsistently defined in the research. In part, this is related to the types of measures used to operationalize advanced dementia. Severity (or stages of dementia) is often measured through functional assessments. Briefly, the Global Deterioration Scale (GDS) stages Alzheimer’s disease specifically into seven stages according to physical and communication impairment, with stages six and seven considered the advanced stage (Reisberg et al., 1984). The Functional Assessment Staging Tool (FAST), a scale based on the GDS, includes additional clinical characteristics in stages six and seven, such as the inability to sit independently, to smile, and to hold the head up (Reisberg et al., 1984). The Bedford Alzheimer Nursing Scale-Severity Subscale (BAN-S) includes ADL items, speech, eye contact and pathological symptoms such as sleep-wake cycle disturbances. The scale scores out of 28, with higher scores reflecting greater severity (Volicer, Hurley, Lathi, & Kowall, 1994). Finally, the Cognitive Performance Scale (CPS; Morris et al., 1994) derived from the Resident Assessment Instrument Minimum Data Set (MDS 2.0)2 includes items related to memory, decision-making, making oneself understood, presence of coma and eating dependency.  While there is overlap in terms of what is included in the scales, there is variability in the research literature in terms of how advanced dementia in end-of-life research is operationalized. For example, while a CPS score of 5 indicates severe impairment, others have defined severe as a stage 6 (Mitchell, Teno, Intrator, Feng, & Mor, 2007), or a stage 4 (Gruber-Baldini, Zimmerman, Mortimore, & Magazine, 2000).                                                 2 The MDS 2.0 is a standardized assessment of function and cognition conducted quarterly on all nursing home residents. It is a mandated assessment in the U.S and in many Canadian provinces, including B.C. It is also widely used internationally. 15    Likewise with the GDS/FAST: Ahronheim, Morrison, Morris, Baskin and Meier (2000) define advanced dementia as per the FAST stage 6d, while others have defined advanced dementia as per stage 7c (Formiga et al. 2007; Giulio et al., 2008) or the GDS stage 7 (Mitchell et al., 2006). This variability contributes to the challenge of making cross-study comparisons, as the population under study is not homogeneous. For example, the illness experience is likely to be quite different for someone who is still able to walk (as per FAST stage 6d) as compared with someone who is reliant on care staff for locomotion (as per FAST stage 7c or higher).  While the aforementioned measures of the severity of dementia are commonly used, it is important to note that the BANS-S, GDS, and FAST were developed in the context of Alzheimer’s disease and not based on observations of other types of dementia. Thus, it is possible that individuals with other types of dementia do not experience cognitive and functional progressive decline as detailed in the scales. However, Ahronheim and colleagues (2000) argue that the FAST describes the functional status of those individuals in the “most advanced stages of various types of dementia” (p. 266) and therefore has utility in operationalizing advanced dementia. Moreover, measures of severity assume an orderly progression of the disease. However, this might not be the case for all individuals with dementia (Lyman, 1989; Small et al., 2007; Volicer et al., 1994). Finally, measures of severity do not consider the complex interplay of contextual factors such as the physical environment (Davis, Byers, Nay, & Koch, 2009) and social positioning (Kitwood, 1997; Sabat, 2006) that contribute to the lived experience. 16    2.1.2 Defining Nearing the End of Life Cox and Cook (2002) have identified three ways of dying with dementia. First, there are those who reach the end of their life with symptoms of physical and cognitive impairment, but are in earlier stages of the disease. Second, there are those who have dementia; however, death is a result of a comorbid condition such as cancer. Third, there are those in the advanced stage of dementia who die as a result of the complications of the disease. This research is concerned with this last category.  People with advanced dementia are frequently not recognized to be dying, which has significant implications for appropriate and adequate support at the end of life. But, what is dying exactly? When is someone identified to be dying and how is care responsive to the needs of an individual who is dying when the boundaries between life and death are not well understood? As Gawande (2014) writes,  [H]ow do you attend to the thoughts and concerns of dying when medicine has made it almost impossible to be sure who the dying are? Is someone with terminal cancer, dementia or incurable heart failure dying, exactly (p.157)?  The uncertainty reflected in Gawande’s questions not only speaks to the ways in which modern medicine has shaped definitions of dying and death (Lizza, 2006), but also mirrors the ambiguity in defining end of life in advanced dementia. At the heart of this uncertainty lie several interrelated empirical and conceptual issues. First, it is only recently that dementia has been considered a terminal illness. In a review of the evidence, Wolf-Klein, Pekmezaris, Chin, and Weiner (2007) note the median survival from time of diagnosis to death is approximately five years. Others have estimated a median survival with Alzheimer’s disease at 7.1 years (95% CI 6.7-7.5 years) 17    and for vascular dementia at 3.9 years (95% CI 3.5-4.2 years; Fitzpatrick, Kuller, Lopez, Kawas, & Jagust, 2005). Advanced dementia itself has a 1.3-year median survival rate, and researchers contend that it parallels similar prognoses of other life-limiting illnesses (e.g., metastatic cancer and stage IV congestive heart failure, Fulton, Rhodes-Kropf, Cocoran, Cahus, & Castillo, 2011; Mitchell et al., 2009). Accordingly, individuals with advanced dementia “can be considered dying” (Blasi, Hurley, & Volicer, 2002, p. 57).  The second issue is the identification of dementia as a cause of death. Only in the last 20 years or so has dementia been included as a cause of death on death certificates in the United States (Gauthier, Leuzy, Raicini, & Roster-Nets, 2013). Thus, while the U.S National Centre for Health Statistics ranked Alzheimer’s disease as the fifth leading cause of death, it remains underreported on death certificates, not only as an immediate cause of death, but also as a condition contributing to death (Watcherman, Kiely, & Mitchell, 2008; WHO 2012).  Third, on the edges of life, dying with dementia is unclear and ambiguous, and there is not a distinct transition point in which the person enters the social category of dying. As Bern-Klug (2004) so aptly observes,  Our social construct of the ‘end of life’ excludes many of (sic) deaths that occur, in part because the dying role is juxtaposed against the sick role, with little formal recognition that most dying now occurs in the context of advanced chronic illness. We avoid admitting that a dying process is taking place until death is upon us (p. 55).    Contributing to this ambiguity is the radical and unerring slowness of death’s approach; the dementia disease process “brings death into life slowly” (Kaufman 2006, p. 29). The illness trajectory of advanced dementia is described as a steady loss of functioning and long periods of disability, with episodes of exacerbated physical 18    symptoms and recovery, until a fatal episode brings about the person’s death (Lynn & Adamson, 2003). While survival is poor after the onset of swallowing challenges, infections, and skin issues (Mitchell et al., 2009), the person “could keep on ‘living on thin ice’ for some years or die in a week” (Lynn & Adamson, 2003, p. 6-7). The consequence of an uncertain trajectory is prognostic inaccuracy. Mitchell, Kiely, and Hamel (2004) observed that while 1.1% of their sample of individuals with advanced dementia was predicted to die within six months, 71% actually died within that period. What often results is the failure to formally identify individuals with advanced dementia as being at risk for death (Chang & Walter, 2010); thus, access to formal end-of-life care is limited.  The fourth issue centres on Western society’s relationship to dying and death and the shifting definitions of death itself (Lizza, 2009). Gadamer (1993) argues that as there is a “systematic repression of death” (p. 63), death and care for the dying are pushed to the margins of public life. This results in the propensity for the public to not engage in the discourse around dying and death, to be ambivalent towards, and to deny death (Becker, 1973). Without a societal-level acceptance that death is indeed part of life, efforts to implement widespread change in care for dying will likely be thwarted. The stigma attached to aging, dementia, and dying puts older adults and persons with dementia at risk for marginalization and being mistreated (Ballanger, 2006; Post, 1995). Coupled with death-denying attitudes these perceptions are likely to translate into care practices that negatively impact care for persons at the end of life (Shemmings, 1996; Komaromy et al., 2000). 19    2.1.3 What is End-of-Life Care? The above section suggested that as prognosis of dementia is characterized by uncertainty, access to end-of-life care is limited. But what is end-of-life care and where does end-of-life care take place? Like the term advanced dementia, the terms palliative-, end of life-, and hospice care are used somewhat interchangeably. However, important distinctions between them exist, particularly as they relate to the B.C context. First, hospice3 relates to a place where end-of-life care occurs, with an admission being for a short term (six months or less). A palliative approach to care refers to a service that targets those who have foregone curative interventions for their illness. This service can be delivered at home, in a hospice, acute care hospital, or residential care facilities, or wherever the person is located. A person can receive a palliative approach to care for several years, though typically they cannot access certain types of benefits,4 which are limited to those with a prognosis of six months or less. End-of-life care is used interchangeably with palliative care and it appears there is little distinction between the terms. For the purposes of this thesis, the term “end-of-life care”5 will be used. This is defined according to the WHO (2015) definition of palliative care explicated on page seven. However, given that dementia is an incurable disease that ends in death, it is somewhat surprising that people with dementia are underrepresented within the                                                 3 It is important to note that the term hospice in literature from the U.S. refers to a program and not a place specifically. 4 For example, B.C. Palliative Care Benefits: http://www2.gov.bc.ca/gov/content/health/practitioner-professional-resources/pharmacare/prescribers/plan-p-bc-palliative-care-benefits-program 5 I used the term end-of-life care because the participants in this study understood the term palliative care differently than the WHO definition. 20    population of those formally receiving palliative care (Small et al., 2007; Sampson 2010). While researchers have advocated for a palliative approach in advanced dementia care (van der Steen et al., 2014), barriers related to prognosis and the lack of recognition that dementia is a terminal illness (Wolf-Klein et al., 2007) persist. Moreover, conversations about dying and death that will assist individuals, families, and care providers to understand the nature of the illness are essential (Caron, et al., 2005; Forbes, Bern-Klug & Gessert, 2000; Small et al., 2007). It is also imperative to have improved end-of-life care education for care providers (Raymond et al., 2013) and support for care providers to navigate the ethical terrain around end-of-life decision-making and the legal and regulatory environment (Davies et al., 2013). While there is a dearth of scientific evidence for the efficacy of palliative care in dementia (Hughes, Jolley, Jordan, & Sampson, 2007), a recent Delphi survey identified best practice and priorities concerning management of physical health and support for families and informal caregivers in decision-making (van der Steen et al., 2014). Sixty-four health professionals and researchers who specialized in palliative and/or dementia care from 23 countries evaluated a set of 11 domains and 57 recommendations related to policy, practice, and research. Overall rating for the importance of the set recommendations observed the following (listed in order of importance): optimal treatment of symptoms and providing comfort; person-centred care, communication, and shared decision-making; family care and involvement; societal and ethical issues; avoiding overly aggressive, burdensome, or futile treatment; education of the health-team; psychosocial and spiritual support; continuity of care; setting care goals and advance care planning; applicability of palliative care; prognostication and timely 21    recognition of dying. Overall consensus was achieved for nine domains and recommendations. However, only moderate consensus was achieved on the recommendation related to nutrition and hydration (under the domain of aggressive treatment) and palliative care in different stages of dementia (though there was a tendency to agree on the applicability of palliative care in severe dementia). Yet, as will be discussed in the following sections, suboptimal treatment of physical symptoms (e.g., pain), overly aggressive treatment, lack of attention to psychosocial and spiritual needs, challenges in advance directives and setting goals of care, and family involvement persist in practice. Clearly, much work needs to be done in order to provide quality end-of-life care for those with advanced dementia. 2.2 The Experience of Physical Symptoms  The characterization of the progression of dementia as a slow deterioration of the body and a dwindling towards death is mirrored in the topics of research that explore the person’s experiences. The bulk of end-of-life research in advanced dementia focuses on what physical symptoms are most present and what medical interventions are implemented to meliorate symptom burden. In this section, I first look at the research that examines symptom burden at the end of life and subsequent medical intervention. I then present a more detailed discussion on two symptoms: pain and swallowing difficulties.  2.2.1 Symptom Burden at the End of Life Over the past couple of decades, a substantial body of research has examined the physical experiences of those with advanced dementia nearing the end of life. This 22    research has been instrumental in informing the clinical trajectory of advanced dementia as well as understanding physical symptoms typically observed as the individual nears the end of life. This area of research owes much to the efforts of Mitchell and colleagues from the United States. Coming from a biomedical perspective, Mitchell et al. have developed a comprehensive research program that seeks to understand end-of-life symptoms for individuals with advanced dementia living in nursing homes environments (Mitchell et al., 2009). Defining advanced dementia as either a CPS of 5 or 6 or a FAST score as stage 7, the researchers tend to use MDS 2.0 data. For this review, I draw upon their more recent prospective study that followed 323 residents with advanced dementia from 22 nursing homes in the U.S. over an 18-month period (Mitchell et al., 2009). Data was collected once per quarter and for those who died (54.8%), data were also collected 14 days post-death via interviews with proxies. The researchers found shortness of breath, pain, pressure ulcers, agitation, and aspiration to be the most frequent distressing symptoms, which increased significantly as the resident neared death. For example, the prevalence of aspirations increased from under 10% to almost 40%, and the prevalence of pain rose from just over 10% to over 25%. Interventions with unclear benefit were also observed; the most common intervention was parenteral therapy6, followed by hospitalization, emergency room visit, and the insertion of a tube feed. However, burdensome interventions were not observed if proxy decision-makers were aware of the resident’s poor prognosis. These results corroborate an earlier retrospective study                                                 6 Parenteral therapy is defined as follows: “Intravenous or subcutaneous hydration or administration of intravenous or intramuscular antimicrobial agents” (Mitchell et al., 2009, p. 1531). 23    (Mitchell, Morris, Park, & Fries, 2004) that used MDS 2.0 data from 2730 residents with advanced dementia. A Dutch study also observed a high prevalence of distressing symptoms over the last week of life (Hendriks et al., 2014). Unlike Mitchell et al., (2009), the researchers found significantly more pain (52% vs. 25%), but marginal differences in shortness of breath (35% vs. 32%) and agitation (35% vs. 33%). Between-study differences can be equated to the difference in time frame (last week of life vs. last 3 months), as well as by the difference in measures used. Whereas the Dutch study reviewed physician after-death reports, Mitchell et al. (2009) used MDS 2.0 data based on observations of residential care nursing staff, reflecting possible observer bias. Finally, Aminoff and Adunsky (2006) examined suffering in end-stage dementia. The researchers developed the Mini Suffering Status Exam (MSSE) and observed 71 individuals with advanced dementia admitted to a geriatric ward over a six-month period. Suffering was operationalized as: not calm, screams, pain, decubitus (i.e., pressure) ulcers, malnutrition, eating disorders, invasive action, unstable medical condition, and suffering according to medical and family opinion. The researchers observed a significant increase in the level of suffering from admission to the last week of life. During the last week of life, the researchers observed that a large majority were not calm, had pressure ulcers, had an unstable medical condition, were malnourished and had eating difficulties, went through invasive medical procedures, and were suffering according to staff and family. While this research aligns with Mitchell et al.’s (2009) study in identifying considerable symptom burden at the end of life, some limitations are noteworthy. Most significantly, the term “suffering” is unclear and thus subject to a wide variety of 24    interpretation. Moreover, as a measure of suffering the MMSE does not appear to capture the multi-dimensional nature of suffering (Boston, Bruce, & Schreiber, 2011). Second, the study population was derived from a geriatric hospital and comparisons to nursing home residents cannot be made, as the illness experience might be significantly different between settings (e.g., it is possible that a person is admitted to the hospital because their condition is unstable).  2.2.2 Responding to Symptom Burden While it is evident that people with advanced dementia experience distressing physical symptoms as they near the end of life, an important question to consider is how health care providers respond to these symptoms. Mitchell, Morris et al. (2004) and Mitchell et al. (2009) observed interventions that they considered to be non-palliative, such as hospitalizations and tube feeds, as well as fewer referrals to hospice. Conversely, Tschirhart, Du and Kelley (2014) report that in the last six months of life, those with Alzheimer’s disease have lower odds of undergoing an intensive procedure (e.g., CPR, gastronomy tube insertion, hemodialysis, enteral or parenteral nutrition) than those with other chronic illnesses.  However, it has been suggested that these medical interventions do not necessarily improve survival or quality of life for those with advanced dementia. For example, feeding tubes do not necessarily improve symptoms associated with swallowing problems and poor nutrition, nor are they associated with improved survival (Mitchell, Kiely, & Lipsitz, 1997; Palecek, et al., 2010). Moreover, health outcomes are not necessarily better following hospitalization (Mylotte et al., 1998) and transfer to a 25    hospital can result in adverse events such as increased confusion, falls, and greater functional decline, often managed by aggressive medical intervention (Ahronheim, Morrison, Baskin, Morris, & Meier, 1996). Antibiotic use is also contested, though research in this area is somewhat contradictory. While an early study found that antibiotics to treat pneumonia offered no improvement in survival or comfort (Fabiszweski, Volicer, & Volicer, 1990), more contemporary research has found antibiotics to improve comfort (van der Steen, Pasman, Ribbe, van der Wal, & Onwuteaka-Philipsen, 2009), or to improve survival but increase discomfort (Givens et al., 2010). In all respects of treatment that responds to symptom burden, consensus points to person-centred approaches that individualizes care according to the needs and preferences of the person with advanced dementia (van der Steen et al., 2014).  2.2.3 Experience of Pain, Expressions of Pain, and Management of Pain The experience of pain has been related to depression, sleep disturbances, anxiety, agitation, and resistance to care (Zieber, Hagen, Armstrong-Esther, & Aho, 2005). In a study of 277 nursing home residents, those with severe dementia experiencing chronic pain were more likely to express combative behaviour, agitation, repetitive behaviours and/or socially disruptive behaviour than those with mild or moderate dementia (Cipher, Clifford, & Roper, 2006). Similarly, vocal perseveration, tense body language, sad facial expressions, and vocal outbursts were perceived to be indicative of pain and discomfort in 104 residents with severe dementia (Kovach, Weissman, Griffe, Matson, & Muchka, 1999). Comorbid conditions such as arthritis, cancer, or cardiac disease (Ferrell, Ferrell, 26    & Rivera 1995) as well as hip fractures, falls, and clinical depression are strongly associated with pain for those with cognitive impairment (Proctor & Hirdes, 2001). While pain is often cited as being ubiquitous in people with advanced dementia, the prevalence of pain is not fully understood. In nursing homes, prevalence rates have been observed at 25% (Mitchell et al. 2009), 31.5% (Proctor & Hirdes, 2001), 47% (Zwakhalen, Koopmans, Geels, Berger, & Harmer, 2009), 52% (Hendriks et al., 2013) and 62% (Ferrell, et al., 1995). While differences in prevalence rates can be attributed to differences in measures used,7 differences can also be explained by the reliance on observers’ judgement and sensitivity to the ways in which individuals with advanced dementia communicate their pain, in the absence of verbal capacities (Herr, Bjoro, & Decker, 2006; Husebo et al., 2007). Moreover, pain is a complex phenomenon influenced by personal perceptions and cultural traditions of both the person expressing the pain and the person assessing the pain (Malloy & Hadjistavropoulous, 2004).  However, suffering from untreated pain can lead to worsening functional impairment, and impaired sleep, and detract from well being (Hendriks et al., 2014). As Johnson (2005) wrote, “unrelenting pain can interfere so completely with thought, self awareness, emotional engagement, and social relationships that it can rob the individual of the experience of being human” (p. S39).                                                  7 For example, Proctor and Hirdes (2001) used the pain scale derived from the MDS 2.0 assessment. This scale was validated with post acute care nursing home residents who were able to communicate and as such, it tends to underreport pain in those with severe cognitive impairment (Cohen-Mansfield, 2004; Teno, Weitzen, Wettle, & Mor, 2001). Conversely, Zwakhalen et al. (2009) used the Dutch version of the pain assessment checklist specifically designed for seniors with limited ability to communicate (PACSLAC-D).  27    Yet, an important body of evidence suggests that pain remains undermanaged. Studies show that the likelihood of being treated with opioid analgesics for pain was significantly lower for those with dementia compared to those who were cognitively intact (Chambaere, Cohen, Robijn, Bailey, & Deliens, 2015; Morrison & Sui, 2000; Kaasalanien et al., 1998; Husebo et al., 2008). Conversely, in a study that examined medication use in 125 residents with advanced dementia near death, researchers observed an increase in opioid medication use as death approached (Blass et al., 2008). Similarly, Hendriks et al. (2014) observed that at least one type of opioid was provided to 73% of residents in pain, and the proportion of residents receiving opioids increased every day in the week leading to the individual’s death. Twenty-one percent of the residents received palliative sedation (the deliberate lowering of consciousness).  Disparate results can be explained by unaccounted-for differences in comparison groups. It is certainly possible that the experience of pain for someone with dementia is different than for someone with cancer. Moreover, it is not clear if the comparisons include those experiencing acute pain, which can be perceived as different than chronic pain. Further, it could also be likely that those who have the verbal capacity to identify their pain are more likely to alert care providers about their pain and consequently have their pain meliorated. In spite of the differing results, it is clear that several barriers to good pain management for people with advanced dementia exist. First, as persons with advanced dementia have difficulties with the interpretation of pain sensations, in addition to communication difficulties in general, communicating the presence and severity of pain is also difficult (Sachs, Shega, & Cox-Hayley, 2004). Hence, assessing pain is rarely straightforward. While there are a growing number of 28    observational pain scales that assess pain, the assessments are not sensitive enough to detect changes in expressions of pain (Allen, Burgio, Fisher, Hardin, & Shuster, 2005) or the intensity of pain (Husebo et al., 2007; Mentes, Teer, & Cadogan, 2004; Proctor & Hirdes, 2001), and are not specific enough to distinguish pain from other affective distress or discomfort (Jordan, Regnard, O’Brien, & Hughes, 2012). While psychometric testing of these scales has been conducted, further testing is imperative as it has implications for how pain is treated; for example, treating noisy vocalizations with analgesics when the expressions are not related to physical pain may not be the best course of action.  Second, frontline staff have a significant role in the assessment and management of pain. While research in this area is limited, studies have shown that nurses’ lack of knowledge about pain and treatment thereof is a barrier to effective pain management of nursing home residents (Jones, et al., 2004; Morrison & Sui, 2001; Zwakhalen, Hamers, Peijnenburg, & Berger 2007). In a study of knowledge and beliefs of nursing home staff (n=123) regarding pain in individuals with dementia, Zwakhalen and colleagues (2007) found significant knowledge deficits in both pain assessment and treatment, though the nursing staff were very satisfied with how pain was assessed and managed in the nursing home. These results corroborate an earlier study that evaluated an education intervention to improve knowledge and attitudes towards pain in nursing home residents with or without dementia (Jones et al., 2004). Results from this study revealed knowledge deficits in management of side effects, the use of non-pharmacological interventions, drug addiction and dependence, and pain assessment. The researchers did not observe a statistically significant improvement in knowledge after the intervention, though 29    improvements were notable across job category. Moreover, care aides’ attitudes were significantly different from those of RN’s and LPN’s, indicating a higher likelihood of socio-cultural beliefs that could interfere with pain management. Similarly, Zwakhalen et al. (2007) found differences in knowledge between regular nursing home staff and more highly educated nurses (pain specialist trainees).  Finally, beliefs about personhood, aging, and dying also factor into the treatment of pain. Malloy and Hadjistavropolous (2004) argue that under-management of pain is rooted in dualistic mind-body distinctions, which focuses on the person with dementia as an object of care and fails to attend to the subjective nature of being-in-the-world, and that the expression of pain is more than a cognitive appraisal of bodily discomfort. As such, beliefs about personhood, and positioning the person as a subject in relationship, have potential to influence how a person’s pain is treated. To test this supposition, Hunter et al., (2013) developed and tested a measure of beliefs about personhood in dementia and modelled its ability to predict responses to vignettes describing residents with or without dementia displaying agitated behaviour. One hundred and seventy-eight participants (nurses and care aides) chose from three interventions: the use of anxiolytics (antianxiety medication), analgesics, and non-pharmacological interventions. The researchers found that in the context of a dementia diagnosis, care providers were less likely to consider analgesics for a person displaying pain-related behaviours but more inclined to use anxiolytics (i.e., medications to reduce anxiety). However, the researchers also found that stronger beliefs about the residents’ personhood were associated with a greater likelihood to express the intention to use analgesics and non-pharmacological interventions to manage pain and a reduced likelihood of recommending anxiolytics.  30    These results are significant because it is the only study that associates beliefs about personhood in dementia with pain management care practices. This is exemplary as researchers in this review rarely make their theoretical lens explicit or question the assumptions underlying clinical practice. Hunter et al.’s study also raises concern that a pharmacologic intervention to treat agitated behaviours (i.e., the use of anxiolytics) pathologizes the behaviour and fails to address the root of the problem (i.e., pain). However, more research is needed as the opposite could also be true; as mentioned earlier, treating anxiety with pain medications when the underlying problem is indeed anxiety is equally problematic. Coupled with the lack of sensitive measures and ways of assessing pain, the failure to acknowledge and listen to the subjective experience of pain undermines the pain experience, devalues the person, and ultimately leads to pain not being adequately treated or other explanations being explored.  2.2.4 Dysphagia, Eating, and Mealtime Experiences As dementia progresses into the later stages the development of swallowing problems or disinterest in eating “are nearly universal” (Hanson, Ersek, Lin, & Carey, 2013, p. 1692) and considered a natural course of the disease process (Gillick & Volandes, 2008). Indeed, Mitchell et al., (2009) reported that 86% of their sample of residents with advanced dementia had problems with food intake and swallowing. Age- related changes in the mechanisms underlying swallowing functions (e.g., taking longer to swallow) as well as swallowing co-morbidities (e.g., sensory changes, dry mouth, loss of lean muscle mass) increase the risk for dysphagia, which can cause choking or food avoidance (Hanson, Ersek, Gilliam, & Carey, 2011; Ney, Weiss, Kind, & Robbins, 31    2009). Inadequate nutrition and swallowing problems are associated with increased risk for weight loss, dehydration and malnutrition, pressure ulcers and poor wound healing, and aspiration pneumonia, and are associated with six-month mortality (Gillick, 2000; Mitchell et al., 2009).  Perhaps the most contested intervention to mitigate these risks is the provision of nutrition through a nasogastric tube or percutaneous endoscopic gastronomy (PEG) tube (collectively known as feeding tubes). Factors associated with the insertion of feedings tubes in individuals with advanced dementia vary considerably across different settings of care (Mitchell, Kiely, & Gillick, 2003; Mitchell, Teno, Roy, Kabumoto, & Mor, 2003), cultural and racial groups (Mitchell et al., 2003; Clarfield et al., 2006), and between countries (Clarfield et al., 2006). However, research has demonstrated that nutrition through feeding tubes does not necessarily reduce symptoms related to poor nutrition and hydration, such as aspiration pneumonia and pressures sores, or prolong life (Mitchell, et al., 1997; Palecek et al., 2010; Kuo, Rhodes, Mitchell, Mor, & Teno, 2009).  As swallowing difficulties emerge, the question is not about the cessation of feeding assistance, but rather, whether there is an alternative to feeding tubes. In spite of relatively high numbers of people with advanced dementia who have feeding tubes (Mitchell et al., 2003; Clarfield et al., 2006), given the evidence, researchers advocate for oral feeding interventions as a way to mitigate the effects of inadequate nutrition and dehydration. Effective interventions include: high calorie supplements, modified foods (e.g., pureed food), enhanced dining environments, appetite stimulants, and careful hand feeding (Hanson, et al., 2011; Ney, et al., 2009; Palecek et al., 2010; Simmons et al., 2008). Paleck et al. (2010) argue for a directive called “comfort feeding only” where an 32    individualized plan of care documents the resident’s signs of distress, what behaviours indicate that she is safe to feed, feeding techniques, and food preferences. The care plan also documents guidelines for cessation of oral feeding, such as, for instance when the resident shows signs of distress. However, in the current rushed and task-focused environment of mealtimes (Hung & Chaudhury, 2009), often with inadequate staffing and lack of staff supervision (Kayser-Jones, 1997), how comfort feeding is taken up in practice is yet to be observed. Critical issues related to forced feeding or insufficient intake related to hand-feeding have yet to be explored. 2.3 Social and Emotional Needs  Thus far, the research that explicates the experience of dying from advanced dementia is primarily oriented from a biomedical perspective, with an unsparing focus on emergent distressing physical symptoms, and the management thereof. Accordingly, it is telling that the experience is characterized as “a state of complete helplessness… the life of end-stage patients with dementia is filled with grief, secretion and stench, suppuration and wounds, crying, screaming or silent pain” (Aminoff, 2012, p. 477). As reflected in Aminoff’s quote the presence of distress, agitation, and challenging behaviours are not uncommon in advanced dementia (Mitchell et al., 2009). Yet, these expressions can also point to affective, spiritual, and social domains of a person’s experience, as well as the potential for these domains to be inadequately supported.  33    Research into the psychosocial experiences of those with advanced dementia identified as nearing the end of life is relatively sparse.8 Psychosocial experiences and needs are primarily understood from family perspectives of end-of-life care in general. For example, in a large study of family members representing 1578 individuals who died in hospital, nursing homes, or at home, Teno et al., (2004) examined quality of care in the last 48 hours of life. Quality was defined as the provision of physical comfort and emotional support, shared decision-making, and whether the individual was treated with respect. Half of the family members across settings reported that the dying individual did not receive enough emotional support and held a belief that the individual was not always treated with respect. Similar findings were observed in a contemporary study of 92 relatives of deceased residents with dementia (De Roo et al., 2015). Just over half of the respondents affirmed that the resident died peacefully, with physical and psychological distress in the last week and last month of life moderately and weak-moderately respectively correlated with a non-peaceful death. Psychological distress however was not clearly defined. These two studies underscore that social and affective domains of an individual with advanced dementia’s experience are not adequately supported; yet, it is clear that these domains are important indicators of quality end-of-life care from the families’ perspective. Qualitative research demonstrates that quality is more than providing physical care. Rather, families identify other important dimensions of wellbeing, namely: resident’s safety and security; care staff making an effort to create a sense of belonging                                                 8 This is not to say that research exploring the social and affective domains of people with advanced dementia is absent from the corpus of dementia research. Rather, it is surprisingly absent in research in advanced dementia cast in end-of-life terms. 34    and attachment through relationships with residents and family; care staff supporting family roles (Lopez, Mazor, Mitchell, & Givens, 2013); and individualized care delivered with compassion (Lawrence, Samsi, Murray, Harari, & Banerjee, 2011). However, the reliance on proxy assessments is methodologically problematic with a variety of variables that impact recall and appraisal of a resident’s experience (for example, gender, age, and own sense of well- or ill-being can influence how the family recalls their relatives’ end-of-life experience; Fakhoury, McCarthy, Addington-Hall, 1997). The reliance on proxy reports is also ethically problematic, as it silences the voices of people with dementia as narrators of their own experiences. Nevertheless, the findings are compelling and identify key areas that require further research and clarification. For example, what does it mean that the resident did not receive enough emotional support? What characterizes fostering a sense of belonging, the delivery of compassionate care, or supportive carer-resident relationship? To move forward in providing person-centred end-of-life care as outlined in the WHO (2002) definition, these types of questions are worth pursuing. Much is at stake if they are not explored. Research shows that people with advanced dementia are at risk of social death (Sweeting & Gilhooly, 1997) well before physical death occurs, experiencing social isolation (Perrin, 1997; Nolan, Grant, & Nolan, 1995), and silencing (Simard, 2007). According to Small and colleagues (2007), these maligning processes represent “too little care” (p. 53) in the end of life context.  Perhaps conceptually these are not discussed as the language of end-of-life dementia care circulates around the language of physical loss and deterioration and the management thereof, rather than the language of ability and remaining capacity (Small et al., 2007). In addition, it could also be related to lack of specificity in terms such as 35    comfort, suffering, and being peaceful—terms that appear on end-of-life surveys—and how those completing the survey understand these terms. What is clear is that social and emotional experiences of those with advanced dementia nearing the end of life are poorly understood. 2.4 Spiritual Needs While spiritual needs and support at the end of life have received some attention in the empirical literature, like social-emotional needs, there is a significant gap in the understanding of spiritual experiences, needs, and support.  Godwin and Waters (2009) sought to explore whether people with advanced dementia had views about their end of life, and if so, whether family and care staff were aware of these. The researchers held interviews with 12 triads (resident, family member, and care staff) and conducted participant observations in eight settings (wards, nursing homes, and one hospice). Resident participant criteria specified the ability to speak at least a few words. The researchers found that individuals with advanced dementia were able to talk and express themselves about end-of-life issues, responding to the researchers’ questions with “humour, fellow-feeling, insightfulness and honesty” (p. 270). For example, one resident interviewed said she “hoped ‘to go to heaven’ but was ‘in no hurry’” (p. 272). However, family and care staff often were not aware of these concerns because they never discussed them with the residents.  Similar results were observed in a phenomenological study that explored existential issues of people with dementia nearing the end of life (Albinsson & Strang, 2002). In interviews with 31 nursing home care staff, the researchers explored what types 36    of existential concerns residents with advanced dementia voiced, and how these concerns were supported. The researchers found that, while many residents were unable to voice their concerns on account of profound deterioration, staff gave three examples of the types of concerns that they encountered: being ready to die, questioning the meaning of suffering, and clearly-expressed death wishes. Staff responded to these concerns in three different ways: the concern was brushed off (“you’re not going to die”); it was avoided (“it’s in God’s hands”); and most commonly, it was responded to with love and empathy (e.g., the carer hugged the person).  Though the above studies only included perceptions of residents with advanced dementia who were able to verbalize, both studies are important as they counter the perception that people with advanced dementia have few emotional, spiritual, or social needs as observed in other studies.9 Moreover, it stresses that low (or no) expectation of achieving valid responses to questions about experiences and attitude towards end of life is unsubstantiated. Indeed, as Godwin and Waters (2009) argue, “if we reject them on those grounds, we return to the position where people with dementia are excluded from research altogether” (p. 270), and excluded from meaningful spiritual care at the end of life. While it is known that the support of spiritual needs at the end of life is an important indicator of family satisfaction with care (Daaleman, Williams, Hamilton, & Zimmerman, 2008), very little is known about actual spiritual care practice and how it                                                 9 Sanders and Swails (2009) interviewed 10 hospice social workers about their perceptions of working with individuals with advanced dementia. While respondents strove to accentuate personhood of those with advanced dementia, they also considered their work unrewarding, limited and ambiguous, and considered persons with advanced dementia to have fewer needs than those who were cognitively intact. 37    plays out in the daily lives of residents with advanced dementia in nursing home settings. Two studies coming from the same research group in the Netherlands offer important insights into this topic. Gijsberts, van der Steen, Hertogh, Muller and Deliens (2013) conducted an ethnographic study of end-of-life spiritual care assessment and practice on a psychogeriatric and somatic unit in a Dutch nursing home. Reporting on the practice observed on the psychogeriatric unit, the researchers found that spiritual needs were never a topic in formal conversations between individuals of the care team or a topic identified in care plans, and physicians were cautious to bring up the topic of spirituality for fear of asking the wrong kinds of questions or inadvertently imposing their own beliefs. However, in spite of the lack of formal discussion, the researchers did observe spiritual care in action through the ministrations of the spiritual counsellor, care staff’s careful attention to the “small things” to support family and enhance residents’ comfort as death approached (e.g., being mindful of touch, tone of voice, music, and aromatherapy), and the facilitation of meaningful relationships between residents and family. Finally, using an example of a resident who had troubles eating, the researchers demonstrated how staff constructed residents’ experiences depending on who was communicating with whom. Amongst themselves, nurses constructed the eating problem as the resident’s way to express her readiness to die, as compared to the nurse talking to the physician, where the problem was presented as a swallowing problem due to the progression of dementia. This reframing of the issued downplayed a spiritual concern, possibly replacing it with a physical problem, and as the researchers argue, “bypass[ed] 38    the discussion and collaboration by nurses and the physicians on a possible spiritual need” (p. 684). In a larger study, van der Steen, Gijsberts, Hertogh and Deliens (2014) examined predictors of end-of-life spiritual care as perceived by physicians who coordinated care in 28 Dutch nursing homes. The researcher reported that only 20.8% of residents (40 out of 207) were provided with spiritual care shortly before death. Family satisfaction with communication with the physician and families’ report of spirituality having been important to the residents, regardless of the importance of spirituality of the physician, were predictive of spiritual care. That only one in five residents received spiritual care is concerning. While the researchers suggest that this number may be underestimated due to a possible narrow definition of spirituality (limited to faith-based definitions), the results are interesting in light of Gijsberts and colleagues’ (2013) finding that physicians were fearful of imposing their beliefs onto the residents’ experience, and thus were reluctant to initiate conversations. Thus, relying solely on the physicians’ perspective may have missed a nuanced understanding of spiritual care and how it is practiced. In this sense, Gijsberts and colleagues’ (2013) ethnography offers a deeper understanding of how spiritual care practices arise informally across staff categories.  It is important to note that physicians appear to have a large presence in Dutch nursing homes, spending between 50% and 90% of their time in the nursing home (Gijsberts et al., 2013) and spiritual counsellors are a typical part of the care team (van der Steen et al., 2014). Thus, results are limited to those settings. However, it is possible that spiritual care is provided informally and in the moment by care staff as observed by 39    Gijsberts and colleagues (2013) in other jurisdictions without formal spiritual care providers. While only Godwin and Waters’ (2009) research directly included the voices of people with advanced dementia (though limited to those who could verbalize), the results of the four studies discussed offers important insights into the spiritual/existential experiences and spiritual care practices for those with advanced dementia nearing the end of life. Recognizing that the notion of spirituality poses particular challenges in advanced dementia owing to not only the difficulties in articulating the experiences, but also the abstract thought needed to appraise existential concerns (Cobb, Dowrick, & Lloyd-Williams, 2012; Vachon, Fillion, & Achille, 2009), it is nevertheless clear that people with advanced dementia can receive solace from ritual, touch, presence, and music (Buckwalter, 2003; Teno, Landrum and Lynn, 1997) and benefit from the careful attention to their psychosocial and spiritual comfort (Gijsberts et al., 2013). More research in this area is critical to not only understand the spiritual needs of those who have limited communication, but also to understand the interplay between care providers’ socio-cultural situation, beliefs about personhood, and the moral imperative of the caregiving encounter.   2.5 Advance Directives  Planning future care involves the collaboration of care providers, family, and the person with dementia herself, and can help prevent disagreement concerning ethically contentious issues such as artificial nutrition and hydration (Aupperle, MacPhee, Strozeski, Finn, & Heath, 2004). Decisions are made in accordance with the person’s 40    known wishes (as documented in an advance directive or living will) or based on her known wishes (as verbalized prior to being identified as legally incompetent). Therefore, it is advocated for the person to plan early on in the trajectory, including designating a proxy decision-maker, who is legally sanctioned to make decisions on the person’s behalf when she is no longer able (Hall, 2009). As noted in section 2.2.2 Responding to Symptom Burden, many of the more aggressive interventions are not necessarily equated with better health outcomes or healthcare. Thus, weighing the benefits and burdens of potential treatments is crucial.  The uptake of directives limiting medical intervention is not necessarily that pervasive (Hall, Shroeder & Weaver, 2002; Mitchell, Kiely, et al., 2004; Morrison & Sui, 2000). Researchers suggest that some people are ill at ease with making decisions about their future care (Moorman, Hauser, & Carr, 2009). Poor communication between care providers and family (Caron et al., 2005; Gessert, Forbes, & Bern Klug, 2000-2001) and the lack of agreement and subsequent implementation of a course of action (Travis et al., 2002) are also factors. Socio-cultural factors are a significant influence in making decisions concerning medical intervention at the end of life (Kwak & Haley, 2005). For example, non-Caucasian men were more likely to have feeding tubes than Caucasians (Kuo et al., 2009; Mitchell et al., 2003), residents in a Jewish care facility were more likely to opt for feeding tubes than those in non-Jewish facilities (Clarfield et al., 2006) and Caucasians were more likely to have a living will, an identified proxy, and do-not-resuscitate (DNR) orders than those in other racial categories (Kiely, Mitchell, Marlow, Murphy & Morris, 2001).  41    Yet, making the decision to forgo care is a tremendous challenge. Consider, for example, feeding tubes. Given the symbolic aspects of care as enacted through food and assistance with eating, a “do not tube feed” order can be interpreted as “no care” (Gillick & Volandes, 2010; Palecek et al., 2010). Without conversation about alternatives to feeding tubes (e.g., hand-feeding assistance), and putting in place shared decision-making processes, the likelihood of implementing aggressive treatment is great. Individuals with dementia and family decision-makers need appropriate information along the illness trajectory, but also require support at key transition points to revisit directives to ensure that it is still relevant to the individual’s current situation (Gillick & Volandes, 2010). 2.6 Advance Care Planning and Family Roles in Decision-Making While advance directives offer direction in terms of the individual with advanced dementia’s treatment preferences, the individual’s proxy decision-maker (most typically a family member), is called upon to make health care decisions as they occur. These decisions often arise in times of crisis and require the proxy to make immediate decisions about interventions such as hospital transfers, feeding tubes, and the use of antibiotic therapies. While much progress has been made in terms of educating family members about their role as a proxy (see Arcand & Caron, 2005; van der Steen, Arcand, et al., 2012; van der Steen, Hertogh, et al., 2012), research has shown that families often feel unprepared to make end-of-life decisions, having not discussed end-stage planning with their relative (Luchins & Hanrahan, 1993) and having insufficient knowledge about the course of the illness and potential medical complications (Caron et al., 2005; Engle, Kiely, & Mitchell, 2006; Gessert et al., 2000-2001).   42    Inadequate knowledge appears to be linked to poor communication between family proxies and care providers. An analysis of interview data derived from 28 family members of residents with dementia revealed that the lack of contact and communication with care providers, and the lack of information about advanced dementia and unfamiliarity with death, resulted in emotional burden, confusion, and feeling ill-equipped to make end-of-life decisions (Gessert et al., 2000-2001). Similarly, Caron and colleagues (2005) interviewed 24 family caregivers to understand their experiences in making end-of-life treatment decisions for their relative with advanced dementia residing in a care facility. The researchers found that while a trusting and supportive relationship between family and care providers was considered essential, one of the most significant indicators of family dissatisfaction with the decision-making process was the infrequency of contact with health care professionals. This resulted in the lack of requisite knowledge about the disease trajectory and what to expect when death was near. As a result, family proxies did not feel well-informed to make decisions about treatment interventions.  While these studies illuminate the communicative practices between family and care providers, and the need for appropriate and timely information, several questions remain unanswered: whose responsibility is it to educate the family? At what point along the dementia journey should family decision-makers receive information about the disease trajectory? Some researchers argue that palliative services need to be moved “upstream,” i.e., initiated at diagnosis and followed through into bereavement (Thompson & Roger, 2014). This sentiment is echoed by van der Steen, Onwuteaka-Philipsen, Knol, Ribbe, and Deliens (2013). The researchers argue that “early basic understanding of 43    dementia as a terminal disease may be a prerequisite to providing comfort at the end of life” (para 1).  What is evident is that families perceive their role in decision-making as a distressing but necessary activity (Ashton, Roe, Jack, & McClelland, 2011). Yet without the right information and open lines of communication, families feel the emotional burden of making these life-death decisions. As Teno, Landrum, and Lynn (1997) state, without adequate psychosocial supports “a family member’s memory of their loved one’s death may be marred by a constant battle of advocating for their loved one and ensuring that appropriate palliative care occurs” (p. 28).  2.7  Conclusion and Purpose of the Research This review of the empirical research on end-of-life care for persons with advanced dementia residing in care facilities demonstrates that, despite significant gains in the understanding of the trajectory of the disease, considerable challenges to provide supportive and responsive end-of-life care persist. Research demonstrates that symptom burden, such as pain and swallowing problems, is high and interventions to meliorate these symptoms reflect not enough care (in terms of pain management, or psychosocial care) or too much care (in terms of tube feeds, Small et al., 2007). While palliative care is associated with better care as it counters aggressive intervention and focuses on comfort, much work needs to be done to integrate a palliative approach to care into practice. While barriers to quality end-of-life care are related to the difficulties in understanding the needs and experiences of people with advanced dementia, it also is impacted by 44    caregivers’ (both formal and informal) beliefs about the needs of the individual with advanced dementia and the appropriate treatment thereof.  However, there are considerable gaps in our understanding of important aspects of the individual’s experiences. Research in advanced dementia end-of-life care is dominated by the biomedical understanding of progressive cognitive and bodily deterioration, which inevitably leads to the prioritization of physical care over a biopsychosocial approach to care. Thus, aspects of a person’s experience, such as psychosocial and spiritual experiences and how these are understood and responded to by family and care staff are downplayed (in spite of their importance being recognized; see van der Steen et al., 2014). This relates to the lack of attention to the subjective, lived experiences of individuals with advanced dementia themselves, the lack of attention to the relational aspects of caregiving (including relationships with family), and the limited consideration for contextual factors such as the care providers’ beliefs, the organizational context, and the physical environment.  Accordingly, this study was born out of the concern about the gap between clinical representations of advanced dementia nearing the end of life, and the everyday lived experiences of individuals, their family, and their caregivers. It is clear that the human experience cannot be cleaved from its manifold socio-cultural and linguistic situation. Therefore, the purpose of this research was to closely examine the experiences of people with advanced dementia identified as nearing the end of life and who reside in nursing home settings. Specifically, it sought to elucidate how care staff and family construct and respond to the psychosocial, spiritual, and physical needs of residents with advanced dementia approaching the end of life; how residents, family, care staff, and the 45    researcher influence one another as co-authors of the experiences; how experiences and care practices shift over this time; and how discursive and contextual factors (e.g., organizational policy and physical environment) shape the individual’s experience.   46    Chapter Three: Theoretical Grounding 3.1 A Persistent Perspective The narrative of advanced dementia, as told in biomedical terms, highlights the story of deterioration and decline of cognitive and physical functioning. This narrative shapes the story of the process of dying, the physical and medical care required to support that process, and advance planning individuals undertake to influence that care at the end of their life. The persistence of this storyline in advanced dementia end-of-life research has left little room for other perspectives to influence or even reshape how that end-of-life story is told. It is not surprising that biomedical research has staked the largest claims in end-of-life research, considering that much of clinical practice is geared towards managing frail, potentially medically complex, aging bodies. This criticism of a biomedical focus is not to understate the value of this understanding, nor to question the complexities in the pathological process of dementia. Rather, the criticism centres on the epistemological and ontological limits of this tradition to understand the whole of the illness experience (Kleinman, 1988).  Rooted in positivist philosophies, biomedical research assumes a detached, objective, and disengaged stance by which to know the world (Bernstein, 1983). This detached view presupposes a value-neutral stance, bifurcating the researcher and researched from each other and their respective social, cultural, historical, and political situations. A consequence of this stance renders the person with dementia as an object of study rather than a subject who can meaningfully contribute to the understanding of her illness experience (Dewing 2002; 2007; Lyman, 1989). An implication of the subject-47    object stance for knowledge production is that the perspectives of persons living with dementia have been traditionally excluded from the research process. The underlying assumption is that the progression of dementia erodes rationality and objectivity and as such, their perspective is deemed unreliable and invalid (Dewing, 2002; 2007).  In ontological terms, the understanding of dementia, on which much of the biomedical research is based (Malloy & Hadjistavropolous, 2004) distinguishes mind from body and positions the former as superior over the latter (Johnson, 2007). This thinking leads to a functionalist and atomistic view of the person where human meaning and understanding of the self and others are attributable to a cognitive process and orientation, unyielding to bodily, emotive, or contextual understandings of the self (Johnson, 2007).  As dementia is cast as a disease characterized by progressive loss of cognitive functioning (cognition being the seat of the self), the person’s very nature of being-in-this-world is questioned. Khuse (1999) for example, argues that those with advanced dementia are nonpersons because they “lack the capacity for self-consciousness, rationality and purposive agency, and have no concept of themselves over time” (p. 359). Status as a person is thusly accorded to rationality and memory, all of which becomes increasingly impaired as the disease progresses. This functionalist perspective of “death of a person” fuels the construction of so-called “zombie metaphors” casting the experience of dementia as a “living funeral” and describing the person as the “living dead” (Behuniak, 2011; Taylor, 2008) In concert with an objective, detached gaze, biomedicine’s reductive view of the person herself gives little credence to the subjective experience of the illness, resulting in a construction of an end-of-life experience as an assemblage of physical, cognitive, 48    affective, and behavioural characteristics—a portrayal of a person who is essentially reduced to clinical problems to be solved or managed. This perspective does little to enhance our understanding of the experience of dying from advanced dementia beyond these constructs, nor does it allow researchers to study the person in her everyday situations, her relationships, or the context in which she is situated. 3.2 A Counter-Story As a counter-story to a predominant reductionist view of persons with dementia nearing the end of life, I position myself within the rich theoretical traditions of social constructionism and phenomenology, which has reshaped the dementia discourse in important ways. Countering the storyline of “the loss of self,” from a social constructionist perspective, personhood theories suggest that the self is not a function of cognition, but is created (and enhanced or diminished) through interaction and relationship (Kitwood, 1997; Sabat, 2006). Research grounded in the phenomenological perspective emphasizes, “The world is not what I think but what I live through” (Merleau-Ponty, 1945/2003, p. xviii). This perspective looks very closely at the experiences of people with dementia as lived and emphasizes that the person is capable of perceiving and understanding the surrounding world (see for example Kontos, 2004, 2005; Phinney & Chesla, 2003). Therefore, the person’s assessment of her experience, whether expressed verbally or through her body, is an essential source of understanding and meaning. A situated-embodied perspective (Hughes, 2001; 2009) expands on these ideas and suggests that a person is located not only in place and time but also in a history, a family, a culture, and a socio-political system (Hughes, 2001; 2009).  49    These shifts have reoriented and reshaped the understanding of dementia and of the self beyond the boundaries of the biomedical perspective and locate the person within the context in which she is situated. Thus, understanding is, as Hughes (2011) writes, one in which dementia is seen as a condition that people live with, and thus one which quintessentially involves persons, where this term generates thought, inter alia, about psychosocial space and moral standing, as well as about pathophysiology (p. 14).  The social constructivist, embodied, and situated perspective moves beyond knowledge production that is objective and disengaged from relationship and community, to one that is embodied, situated, and relational. These are important vantage points for understanding the experiences of those with advanced dementia nearing the end of life. However, even though there is a significant body of research that explores personhood, embodiment, and relationship of individuals in earlier stages of the illness, the voices of those with advanced dementia nearing the end of life (in advanced stages of the illness or otherwise) are largely absent. In part, this reflects a methodological problem. When we talk about understanding a person’s experience from her perspective, how can that perspective be solicited when the person’s capacity to communicate significantly diminishes?10 Are another person’s reflections of a person with dementia’s experience legitimate (i.e., express the nature of another’s experience)? That is, operating under the assumption that we live in a shared world, are the person with dementia’s experiences accessible and not alienated from others around them? This question highlights the underlying problematic in my guiding research question. If I am curious about how carers                                                 10 In this statement, I am assuming significantly reduced verbal capacity as the illness progresses (as per Reisberg et al.’s (1982) observation), and I am also assuming reduced embodied cues (for some) as the body becomes increasingly frail. 50    and family construct the experiences of people with advanced dementia nearing the end of life, are these constructions reflective of the person’s experience and what do they say about the person’s everyday lived reality? These questions ontologically locate the nature of being (selfhood) in relationship with others and within the context in which the person is emplaced, and epistemologically assumes that understanding of experience (i.e. the construction of knowledge) is co-produced within that relational situation.  In what follows, I draw upon the dialogical perspective as threaded through the writings of the German philosopher Martin Buber, the Russian philosopher and literary critic Mikhail Bakhtin, and researchers and theorists who have been inspired by their work. I first orient the reader to the underlying ontological assumptions of a dialogical perspective. I then discuss specific concepts that are relevant to the understanding of the shared nature of experience and therefore to understanding the construction of experience of people with advanced dementia nearing the end of life. 3.3 A Dialogical Perspective A dialogical perspective rejects dualist separations of self/body and self/other and claims that human existence and human meaning are created within and by relationships (Cresswell, 2011; Marková, 2003; Taylor, 1991). Buber’s (1923/1970) and Bakhtin’s (1963/1984; 1981/1992) understanding of the self is wrapped in a sense of the other. That is “a person has no internal sovereign territory and that he is wholly and always on the boundary with others” (Marková, 2003, p. 256). Self is dynamically situated and created within and by relationship and thus cannot be conceptualized as singular or unified but as “co-being” simultaneously with other beings (Cresswell, 2011; Hess, 2003; Marková, 51    2003). As such, everyday reality is shaped through interactions between social actors, and phenomenological lived experiences are deeply entwined with the perspectives that shape them (Cresswell & Teucher, 2011). While a social actor operates from a unique perspective, meanings (and interpretations) attached to events, behaviours, experiences, and ideas are constructed through a process of coming to a shared understanding, which necessarily involves interrogating one’s perspective in relation to another (Bakhtin, 1981/1992; Cresswell, 2011).  The socially-constituted self offers a counter-argument to the notion that selfhood is lost due to the progression of dementia (Cohen & Eisdorfer 1986; Davis, 2004; Khuse, 1999), as selfhood is intimately tied to another person. For example, drawing upon the relational perspective of Buber (1970), Thomas Kitwood (1997) theorized that relationship is central to the notion of personhood in dementia. Defining personhood as “a standing or status that is bestowed upon one human being by others, in the context of relationship and social being…implying recognition respect, and trust” (p. 8), Kitwood posited that the caring relationship can support or malign personhood. As such, while relationship between social actors has ontological significance, it also has moral implications for how social actors position one another. 3.3.1 Co-Authorship: The I, the Thou, and the Space Between In the beginning is the relation (Buber 1923/1970, p. 69).  In his classic work “I and Thou,” Martin Buber (1923/1970) distinguishes between two ways of being-in-relation. The first is termed I-Thou relationships, which are characterized as mutual and open relationships that require commitment and involvement. The second is termed I-It relationships, or subject-object relationships in 52    which one relates to another in a detached manner and without any moral commitment to foster a caring relationship or accentuate selfhood. The other person is seen as a means to an end. As argued by Kitwood (1997), I-It relationships have a sobering history in dementia care with little evidence of close and open relationships that morally position the person with dementia as a “Thou.” Maligning care practices (e.g., stigmatization, invalidation, and objectification) and assumptions that the disease of dementia erodes all capacity, agency, and thus selfhood do very little to value the person.  However, Buber does not set up these two ways of being with each other as an either/or. Rather, the nature of relationships is recursive, as Westerhof, van Vuuren, Brummans and Custers (2013) succinctly write, “An I who treats another person as an object (It) becomes an It, just like treating another as a Thou transforms oneself into a Thou” (p. 356). Thus, how carers relate to the person with advanced dementia is not only essential to the person’s selfhood, but also to their own. Moreover, Buber accords significance to the space between, where I-Thou and I-It is best understood as a back-and-forth movement between the two, a continuous alteration to reduce the distance between the I and Thou. Buber’s close examination of this essentially moral relationship operates primarily at the interpersonal level. Owing to Bakhtin’s broad understanding of dialogue, his treatment of the I-Thou relationship is more expansive than Buber’s (Marková, 2003). Bakhtin’s insight into the literary craft of Dostoevsky (Bakhtin, 1963/1984) brings to light the I-Thou relationship as manifested between author and character; namely that a character is not determined by the author, but is formed in ongoing interaction (i.e., dialogue) with other characters, including the author. While Bakhtin expresses his ideas 53    of the I-Thou relationship as co-authorship and co-participation (Cresswell & Teucher, 2011), he focuses on the interplay between multiple voices11 located in interpersonal relationships, in community, culture, a political environment, and social institutions. This interplay, as Marková (2003) argues, “demands evaluation of the other, struggle with other and judgment of the message of the other” (p. 256). That is, the self seeks to make meaning of the speech and thoughts of others by asserting her own position and/or appropriating another’s perspective into her own understanding. This speaks to how one positions oneself and the other in relationship12 and underscores that while the “I” and “Thou” exist in a shared relational space, there is an important tension that exists in the “space between us” (Buber, 1923/1970; Holquist, 1990). This space is negotiated, partial, and never fully transparent and knowable (Davis & Harré, 1990; Frank, 2005; Holquist, 1990). In the context of the I-Thou relationship between a person with advanced dementia and her carers, the space is characterized by asymmetry that arises not only from the imbalance in communication resources, but also from power imbalances inherent in carer-care receiver relationship and the context within which they are situated (e.g., social institutions and political environment). Part of the ethical task then is to work the in-betweens (Fine & Weiss, 1996), finding places of intersection and resonance between the I and Thou and seeking ways to minimize the asymmetry and reduce the distance between (Buber, 1923/1970; Marková, 2003). Central to this ethical task is that                                                 11 The notion of ‘multi-voice’ is essential to the understanding of the I-Thou relationship and the tensions that arise in seeking to understand the other person in relationship. This will be further explored in the next section. 12 Davis and Harré (1990) and Harré and van Langenhove’s (1991) positioning theory is relevant to this discussion as it highlights the tensions that Holquist refers to, namely that positioning in interaction is not consistent or necessarily intentional, but is dynamic and always relative to the other.  54    the I-Thou relationship involves constant and active negotiation. The key words here are “constant” and “active,” which implies that those in the relationship are never finalized or fully determined. “As long as a person is alive he lives by the fact that he is not yet finalized, that he has not yet uttered his ultimate word” (Bakhtin 1963/1984, p. 59). Frank (2005; 2009) argues that the notion of unfinalizability is central to research praxis, writing that “all that is unethical begins and ends when one human being claims to determine all that another is and can be” (Frank, 2005, p. 966).13 In what follows, I bring to the fore three key constructs that characterize the I-Thou space: utterances and the notions of multi-vocality and dialogical tension; embodiment and intercorporeality; and, emplacement and situatedness. 3.3.2 Key Constructs of a Dialogical Perspective 3.3.2.1 Utterances, Multi-vocality and Dialogical Tensions Every individual lives in the world of others’ words (Bakhtin,1979/1986, p. 143).   Foundational to Bakhtin’s understanding of a dialogical perspective is his understanding of how we communicate with one another. The essence of dialogue as reflected in the space between the I and Thou is the simultaneous difference from, yet merging with, another. In what follows, I describe several key components of what occurs in dialogue. First, Bakhtin is clear that dialogue contains not only the words we speak, but also other modes of expressions such as embodied action (Cresswell & Teucher, 2011) and “the ‘languages’ of art, of gesture, of love” (Taylor, 1991, p. 33). These communicative                                                 13 The notion of “unfinalizability” has important implications for research practice and will be noted in Chapter 4, p. 112. 55    signs, or utterances, reflect personal experience and the social context in which they are located. While utterances are oriented and situated in the speaker’s perspective, they do not exist unless directed towards an addressee, whether in a face-to-face situation or with an imaginary other (Bakhtin, 1979/1986; Reissman, 2008). Thus, an utterance is a unique social act, located in the I-Thou relationship where meaning is co-produced, and a multiplicity of perspectives can co-exist. Second, words and gestures do not simply convey one meaning; rather, they are multi-voiced.14 As language is a living situated reality unhitched from its theoretical structures, a word does not relate to its object in a single way, and as such has multiple meanings, layered in ideology, meanings, and perspectives (Bakhtin, 1981/1992). As Reissman (2008) so aptly expressed, “an utterance carries the traces of other utterances past and present, as words carry history on their backs” (p. 107). Thus, an utterance is open to interpretation and multiple understandings are possible.  Third, in addition to the tension noted earlier (i.e., the Thou is never fully knowable), multivocality and polyphony (i.e., multiple understandings) give rise to another important tension. Bakhtin (1981/1992) expresses this tension as opposing tendencies: forces that seek to unify language versus forces that seek differences and multiple ways of understanding (centripetal and centrifugal forces). Monologue is the antithesis to dialogue as it expunges everyday experiences from its dialogical orientation. However, the two opposing tendencies are a necessary part of dialogue as they are key to                                                 14 Bakhtin uses the term “heteroglossia.” I am using the term multi-voice, following theorists such as Reissman (2008), Cresswell (2011) and Marková (2003). 56    understanding the intersubjective dance between self and other, and between unique perspectives and shared understandings.  It is clear that the expansive nature of dialogue is to be taken seriously, as our experience of the world and our relationship within it are shaped by the utterances we employ, and by those utterances directed to us (Bakhtin, 1981/1992). Words and the meanings ascribed to them can create distance. Consider, for example, the term the resident with advanced dementia, a typical term used to describe individuals who live in institutional settings. This term assumes acquiescence to normative roles (e.g., the person as a passive recipient of care) and positions the person with advanced dementia in a generic social category, which can distance a health care practitioner from the person and her lived experiences and promote a power-over type of relationship. In this sense, the term resident with advanced dementia is situated in an authoritative discourse (i.e., the monologization of the person’s experience) that typifies nursing home care, which describes the person in accordance to disease categories and care as task-based. As Bakhtin (1981/1992) argues, authoritative discourse seeks unconditional allegiance, is persuasive, and does not merge with other types of discourse; “it remains sharply demarcated, compact and inert” (p. 343). Ultimately, it promotes a monologue. As such, the term “the resident…” framed within this monologue will encourage a particular way of talking about the person (e.g., according to the mechanisms by which to assist her) and her experiences (e.g., ascriptions of lack of awareness). This contributes to distance and asymmetrical relationships in the I-Thou space. In Kitwood’s (1997) conceptualization of I-It relationships, the notion of monologue is brought to sharp focus. A purely instrumental and objectifying approach to dementia care does not open the dialogical 57    space for a response, essentially quelling the perspective of the person with dementia. Thus, institutional care becomes the script to which all must adhere.  A large part of the research pertaining to advanced dementia nearing the end of life seems to promote this unified way of thinking by accentuating progressive decline and deterioration and the subsequent responses of carers (and healthcare systems) and family. As an authoritative discourse, it is very persuasive, defining and describing in biomedical terms the parameters of a person’s experience. However, like describing dementia according to the functional stages described in Chapter 2, the uncritical acceptance of conceptualizing the end of life in those particular terms results in a dissonance between monologic “truths” and the lived experiences of those with advanced dementia nearing the end of life. It finalizes the person, and suppresses the opportunity for her voice, and those who care for her, to be heard.  3.3.2.2 Embodiment and Social Embodiment  A significant piece to the dialogical space centres on the notion of embodiment which has its roots in the phenomenological perspective articulated by Merleau-Ponty (1945/1962) that “the body is our general medium for having a world” (p. 169) and the experience of our being-in-the-world is an experience of being-with-others. This perspective underscores that our perceptions, understanding and identity originate in and are experienced through our bodies. We know, establish meaning, and understand our world through our engagement in it (Johnson, 2007). Bakhtin suggests that the body should be thought of as a social entity as “one enters into unity with others by virtue of being caught up in the ‘people’s mass body’ in enacting a corporeal style along with others” (Cresswell & Baerveldt, 2011, p. 266). That is, a body cannot be conceptualized 58    outside the web of interrelationships of which it is a living part (Holquist, 1990). The commonalities and shared experiences that arise from participation in life with others reveal the socially constructed “bodily living of self” (Cresswell & Baerveldt, 2011, p. 269).   Bakhtin perceives embodiment to be tied to language. That is, through bodily expressions the body communicates something about the person and provides a vehicle for developing a deeper understanding of the person’s everyday experiences (Cresswell & Baerveldt, 2011; Fredriksson, 1999; Hughes, 2001; Kontos, 2004; Wright & Bratjman, 2011). The primacy of the body in relationship to understanding individuals with advanced dementia has been developed by Pia Kontos. Derived from her readings of Merleau-Ponty and Bourdieu, Kontos (2004), developed the notion of embodied selfhood as referring to the complex inter-relationship between priomordial and social characteristics of the body, all of which reside below the threshold of cognition, are grounded in the pre-reflective level of experience, and are manifest primarily in corporeal ways (p. 837). Kontos (2004; 2005) proposes that a person’s embodiment tells a story of this person revealing her social and cultural habits through bodily movement. She has argued that the body is foundational to selfhood as “the body is an active, communicative agent, imbued with its own wisdom, intentionality, and purposefulness, separate and distinct from cognition (Kontos, 2005, p. 555). While Kontos does not reference the dialogical situation, her work echoes a dialogical position whereby the world of experience is simultaneously social and corporeal; our bodily presence cannot be disentangled from the social, historical, and cultural elements within which it is situated (Cresswell & Baerveldt, 2011; Dufva 2004; Holquist, 1990). Moreover, Kontos’ (2004) understanding 59    that “during interactions, words assume a gestural significance” (p. 840) highlights how meaning emerges in social contexts and is not contained in the word alone. Thus, embodiment not only speaks to our own experiences of being with each other in this world, but also enables us to become known to each other (Dufva, 2004). However, while Kontos’ (2004) work accentuates the embodied selfhood, her work pays less attention to how body gestures are understood, that is, how listeners construct meaning and understanding of the person’s communicative cues and their subsequent response (the active dialogue between the I and Thou). Bakhtin (1981/1992) accords significance to the listener. He writes “primacy belongs to the response, as the activating principle: it creates the ground for understanding, it prepares the ground for an active and engaged understanding. Understanding comes to fruition only in the response” (p. 282). Frank (2004) suggests that palliative care is inherently dialogical as it is grounded upon not only how a person may story her or his life, but also in the “capacity of caregivers to hear what their patients say” (p. 111). However, without the response, what is heard may have little bearing on the person’s experience (i.e., a null response). Consider, for example, the assessment of pain. Implicit in pain assessment for people with advanced dementia is that bodily cues (such as facial grimaces) imply something about the person’s experience of discomfort, whether physical pain, or affective distress. However, Hunter et al.’s (2013) research linking carers’ perceptions of personhood to their propensity for interpreting embodied gestures as expressions of pain highlights the significance of the listener and her response. Carer staff who held the perception of embodied selfhood were more likely to recommend treatment for pain; their capacity to hear, and respond to, what the person was expressing was clearly influenced by their 60    perceptions of personhood. However, what this research, and pain assessments in general, are limited by is that they do not give just credence to the dialogical situation and the meaning that arises in the intercorporeal way we are together.   Yet, before a response is given, understanding of the other person must be generated. Merleau-Ponty (1945/1962) suggests that “the other can be evident to me because…my subjectivity draws its body in its wake” (p. 410). Gallagher and Zahavi (2008) write, “When I experience myself, and when I experience others, there is in fact a common denominator. In both cases, I am dealing with embodiment and one of the features of my embodied subjectivity is that it, per definition, entails acting and living in the world” (p. 206).  In the caring sciences, the notion of embodiment has been used to understand the needs and experiences of others through the act of caring (Bergum & Dossetor, 2005; Wright & Brajtman, 2011). The proximity of care and the physical intimacy between the carer and the person with illness affords a particular kind of knowledge. In care for those with advanced dementia nearing the end of life, this intimacy can be encapsulated in the notion of bodywork (Twigg, 2000; Twigg, Wolkowitz, Cohen, & Nettleton, 2011). Bodywork “involves direct, hands-on activities, handling, assessing and manipulating bodies. It is often ambivalent work that may violate the norms of the management of the body, particularly in terms of touch, smell or sight” (Twigg et al., 2011, p. 172). The knowledge that a carer derives from direct experience of the person’s body is essential in how the carer responds to the person. Carers build a stock of knowledge as they observe, engage with, and touch the body of the person over time, and together they build a repertoire of how to be with each other in a relational space. Thus, being in close bodily 61    contact and touching the body of another affords direct knowledge of the person and “as we ‘feel’ the other person…it is difficult to be immune to the effects our actions have on the other” (Bergum & Dossetor, 2005, p. 148). Caring practices that are mindful of the embodied reality of the person shift practice from the task (i.e., the object of a carer’s labour) to bodies in relationship. In addition, care engenders physical intimacy and close proximity, which provides a point of resonance with the other person’s experience by generating empathy, love, care, and concern (Kendrick & Robinson, 2002). Gallagher and Zahavi (2008) suggest that the phenomenological understanding of empathy is much more than making inferences about someone’s experience based on her bodily behaviour. Bodily empathy originates from our shared existence in the physical and social world and therefore is a vehicle by which we experience the other person directly (Gallagher, 2012; Gallagher & Zahavi, 2008; Rudebeck, 2001). However, this does not mean we have “first-person access” to another’s experience whereby  we can experience the other precisely in the same way as she herself does, nor that the other’s consciousness is accessible to us in precisely the same ways as our own…[Instead], It is possible to experience minds in more than one way. When I experience the facial expressions or meaningful actions of another, I am experiencing another’s subjectivity; and not merely imagining it, simulating it, or theorizing about it. The fact that I can be mistaken and deceived is no argument against the experiential character of access (Gallagher & Zahavi, 2008, p. 204). Gallagher and Zahavi’s (2008) reflections enable us to take into account how knowledge of a person with advanced dementia is cultivated in the care context by drawing attention to the immediacy of the everyday reality of the person, social embodiment, and the web of interrelationships in which the person and her carers participate. Moreover, this enables one to carefully consider the dynamics of the dialogical space within which 62    individuals intersect and combine to create a co-constructed space. This is particularly important in the understanding of advanced dementia because the person is so reliant upon others to not only interpret any communicative cues, but also to be an enabling agent, so to speak, to help the person remain connected to the community (Small et al., 2007).  Current discourse in dying from advanced dementia tends to obscure what takes place between bodies (e.g., the person with advanced dementia and the carers) with the more dominant narrative being the technical management of the dying body. As social embodiment reflects the language of our dialogical situation, it not only refers to the immediate dyadic (or triadic) relational space, but also to the “embodied know-how involved in enacting the generic style of a community” (Cresswell, 2011, p. 477). Thus, our attention is also directed towards the nursing home environment and the organization of care surrounding the person with advanced dementia nearing the end of life; treating these as specific “language communities” that cultivate a particular way of being-in-the-world not only for the person, but also for the family and care staff. For example, embodied know-how is influenced by policies and practices that guide end-of-life care. It is also influenced by education and the tacit and acquired skills of care staff.  In turn, embodied know-how contributes to the way dying is experienced by those who constitute the dialogical space. Thus, who and what constitute this language community is an important question to explore. This brings us to the final conceptual issue guiding this dissertation. 63    3.3.2.3 Emplacement and Social Situatedness The final area of the dialogical perspective I wish to draw attention to is the notion of emplacement and social situatedness. Inasmuch as language is embodied and is situated in the here-and-now, it extends beyond its local context by the nature of language being multivoiced. In addition, Bakhtin accords much importance to the context in which dialogical relationships are situated, in terms of the local situation (i.e., the immediate interrelationship) and the broader socio-cultural and political environment. While this is implicit in much of his writing, and how others take up the notion of social embodiment (Cresswell, 2011; Cresswell & Baerveldt, 2011), I foreground the concept to accentuate its importance to the dialogical situation. The notion of emplacement as explicated by Ingold (2011) and Casey (1993) has ontological significance: to be is to be in place. Casey writes, The power a place such as a mere room possesses determines not only where I am in the limited sense of cartographic location but how I am together with others (i.e., how I commingle and communicate with them) and even who we shall become together.  The “how” and the “who” are intimately tied to the “where,” which gives to them a specific content and coloration not available from any other source.  Place bestows upon them “a local habitation and a name” by establishing a concrete situatedness in the common world. This implacement is as social as it is personal (p. 23). As indicated in the quote above, place not only contains the I-Thou relationship, it has the power to shape interaction. In this research, I have conceived the I-Thou to be manifested primarily in the care encounter15 (Twigg, 2000) between the person with advanced dementia and her carers. The care encounter is a way of being-with-others as                                                 15 The term “care encounter” was used by Twigg (2000) to describe care work in the home care situation. I use the term to describe interactions that arise in the nursing home setting and include all interactions between the residents, staff, family and visitors. 64    part of our embodied and emplaced co-existence. Yet, the care encounter does not stand in isolation from where the interaction occurs; that is, it is emplaced within a relationship (as described above), within a specific physical environment, and within a specific care environment, which in itself is located in a larger healthcare system.  The nursing home environment, as a systems-world, is modelled along market principles that focus on routine and standardized practice using objective and measurable tasks (Froggatt, Hockley, Parker, & Brazil, 2010; Small et al., 2007). This results in a care environment that champions practices that “construct the recipients of care as tractable and predictable, transforming their bodies into appropriate objects of labour” (Twigg et al., 2011, p.179). Yet, the care environment is also a place where people live, work, and visit and where lives are entangled through dynamic interaction creating a collective identity (Ingold, 2001; 2011). The people who weave in and out of the nursing home structure bring their own stories, ways of thinking and understanding, and ways of being-in-the-world. This creates a multiplicity of perspectives (i.e., the polyphonic situation). It would come as no surprise then, that the systems-world and life-world of a care facility environment fuel tension. Research in nursing home environments has revealed the preponderance of the systems-world to dictate the life-world, resulting in residents and care staff being subjugated and disempowered by the larger system in which they are located (Diamond, 1986; Lopez, 2006). In end-of-life research in nursing home environments, researchers have observed that losses in physical capacity (e.g., incontinence, immobility), incoherent speech, and severely impaired cognition (i.e., symbols of loss of control) lead to sequestration and social rejection (Chan & Kayser-65    Jones, 2005; Froggatt, 2001; Kayser-Jones et al., 2003). These iatrogenic social processes are compounded by systems issues such as inadequate staffing, lack of staff supervision, and poor physical environments (Kayser- Jones, 2002; Kayser-Jones, et al., 2003), that place carers in an intolerable bind when trying to provide person-centred care at the end of life and ultimately creates distance in the I-Thou relationship. 66    Chapter Four: Path of Inquiry 4.1 Introduction In this chapter, I outline the path of inquiry, detailing how the theoretical and empirical grounding discussed in the previous chapters was realised practically in the field. The chapter focuses on the technical and practical aspects of field research as well as how the process was lived as the research unfolded over time. The first section outlines the epistemological orientation of the dissertation, followed by a discussion of the relevance of ethnographic case study methods as a strategy of inquiry. The next section titled Entry into the Field describes the process of gaining entrée. This is followed by recruitment of study participants and locates the research context. The following section, Data Collection Procedures outlines the data collection processes, namely: participant observations, interviews, video observations, and review of relevant documents. Analytical Procedures presents the approach to data analysis. The final section presents strategies used to achieve quality and trustworthiness. 4.2 A Constructivist Grounding  In light of the dominant biomedical constructions of the experiences of advanced dementia nearing the end of life, this research seeks to shift the locus of understanding from objective decontextualized knowledge to that which is ethically and socially constituted. As explicated in the previous chapter, the theoretical perspective of dialogism highlights how this research assumes that meaning and making sense of experience are constructed, co-constructed, and continually reconstructed within socio-cultural circles of influence (Cresswell, 2013; Frank, 2005; Holstein & Gubrium, 2008). 67    By being attuned to how end of life in advanced dementia is experienced, this research seeks to unearth the elements and organization of constructed realities that make experiences of the social world visible (Holstein & Gubrium, 2008). Motivated by interest in the practical, everyday activities people are engaged with, and how interaction and language construct social reality, it follows that I do not seek to relay fixed and universalistic conceptualizations of phenomena. Rather, I hope to expose the more fluid, particular, and socio-cultural embedded conceptualizations of them, offering a partial understanding and opening up the possibility for alternative readings (Frank, 2012; Weinberg, 2008). As Hacking suggests, the intellectual value of constructionism is to consider the practical point in research and that is to “promote a better way of thinking and, more important, living with respect to the world we inhabit” (as cited in Weinberg, 2008, p. 15) 4.3 Strategy of Inquiry: Ethnographic Case Studies In the study of the experiences of individuals with advanced dementia nearing the end of life, I sought a methodology that supported a commitment to meaning-making and experience; a commitment to ethical inclusion of the individual with advanced dementia and her caregivers; a commitment to the fluid, constructed, and polyphonic nature of an intensive research process; and a commitment to a rendering of a thick description of the person’s lifeworld (Frank, 2005; 2012). In considering these commitments, I used ethnographic case study methods as the strategy of inquiry by which to explore the experiences of people with advanced dementia nearing the end of life.  Ethnographic research is oriented towards an intimate understanding of people’s actions, interactions, and experiences within everyday contexts, with a goal to 68    communicate thick descriptions of the social world (Emerson, Fretz, & Shaw, 2011; Prus, 1996). Within these everyday contexts, ethnographers attend to how social actors make meaning and “work out their situations in conjunction with those others whose lives intersect with their own” (Prus, 1996, p.164). Thus, ethnography offers a methodology that enables careful attention to the study of human group life and the dynamics of human interchanges primarily by achieving intimate familiarity with the nature of the phenomenon under study. From a constructivist perspective, in addition to describing the what of social life, ethnographers are also concerned with “how social realties are produced (Holstein & Gubrium, 2008, p. 374).  Case study research is defined “by its intensive focus on one or more cases purposefully selected for study” (Sandelowski, 2011, p. 156). Ethnographic case studies allow for the study of complex social situations in the real-life context and enable an exceptionally rich means of unearthing insights into people’s lived experiences (Stake, 1995; Yin, 1994). Case study methods have a rich tradition in social and health sciences research and are used to understand a variety of contexts with a variety of objectives; for example, to evaluate, describe, explore, and understand phenomena related to individual or collective experiences and perceptions (Anthony & Jack, 2009). Walshe, Caress, Chew-Graham, and Todd (2004) outline several reasons to utilize case study methods in palliative research. First, the complex and individual nature of dying demands in-depth and robust methods that can “examine and expose that complexity” (p. 678). Related, understanding the phenomenon from multiple perspectives, a key feature of case study design, helps uncover the complex nature of dying. Third, case study methods are particularly useful when the context and situational factors (and the shifting nature 69    thereof) that influence the dying experience are important to the study. Fourth, case study methods offer flexibility in the study design. Design of the study is amenable to adaptation to meet practical, ethical, and theoretical concerns. Finally, case studies often have practical significance. The details that lie in the description of the case study, while context-dependent, have tremendous value for those caring for someone nearing the end of life, and while they cannot be generalized to another context, they can raise awareness of other like-situations. Aligning with ethnographic methodology, this case study emphasizes three core features of ethnography: 1) participant observation, i.e., sharing in the day-to-day activities of those I am studying; 2) participant perspective, i.e., the phenomenon of interest is studied from the perspective of the participants; and, 3) participant emplacement, i.e., the phenomenon is contextualized and referenced to the setting in which it occurs (Ingold, 2011; Prus, 1996). Moreover, aligning with the constructivist perspective, I see the data that I have gathered not as an objective account of the participants’ experiences, but as knowledge co-produced with participants as it pertains to their specific case. In other words, I recognize that my presence as a researcher inherently shapes what is described and, accordingly, necessitates researcher reflexivity throughout all stages of the research process. Case study research is not without its points of contention. Two spaces of tension that arose in this research were differentiating between method and methodology and countering dismissals of this research due to its lack of generalizability. First, in terms of method/ology, I align myself with those scholars who assert that case study research is primarily about "a concentrated focus on a number of cases" (Sandelowski, 2011, p.154), 70    and not a methodology per se with its own clear ontological, epistemological grounding that guides analytical procedures and its narrative structure (Cresswell, 2013; Yin, 2003). Luck, Jackson, and Usher (2006) assert that case studies are a research strategy informed by varied philosophical tenets, theoretical perspectives, and an applied definition of the case that enables flexibility in methods and potential applications of findings.  Regardless of where one sits in terms of method/ology, what is clear is that how the case is defined is of paramount importance (Ragin, 1992; Sandelowski, 2011; Stake, 1995). Cases are bounded empirically, theoretically, and spatially (Luck et al. 2006; Ragin, 1992; Sandelowski, 2011) and are constructed by the researcher for their informational representativeness (Sandelowski, 2011). Ragin (1992) also differentiates between a case that has been designated as a specific instance of the phenomenon and a case that is general and relative external to the conduct of research. As will be discussed, this case study was bounded empirically, temporally, spatially, and theoretically; it also is “specific,” in Ragin's (1992) term, as it represents a particular phenomenon (a person with advanced dementia's experiences at the end of life). Bearing in mind that the case is a specific instance, though it may be regarded as a typical case, it is important to note that it cannot be generalized to other cases (Stake, 1995). This is the second space of tension. In describing my research to others (lay persons, academics, and scientists alike), it was common to receive criticism for the apparent lack of generalizability with a sample size of two. In its classic form, generalization is related to universals and the claim that findings are temporally and contextually fixed and absolute (Lincoln & Guba, 2009). However, a more contemporary understanding of generalization underscores the notions of fittingness (i.e., the degree of 71    congruence between contexts; Lincoln & Guba, 2009), analytical generalizability (i.e., where the researcher strives to theoretically align results; Yin 1994), and naturalistic generalization (i.e., adding to a reader’s understanding through resonation; Stake, 2009). Flyvberg (2006) writes, “Formal generalization is only one of many ways by which people gain and accumulate knowledge. That knowledge cannot be formally generalized does not mean that it cannot enter into the collective process of knowledge accumulation” (p. 227). Flyvberg (2006) underscores the epistemological value and the power of examples and as Stake (1995) argues, “The real business of a case study is particularization, not generalization. […] the first emphasis is on understanding the case itself” (p. 8). Thus, what might be lost due to lack of generalizability is balanced by the specificity and significance of obtaining highly detailed knowledge regarding an area of concern, namely an understanding of a particular case's experiences of advanced dementia nearing the end of life and how her life is understood by those who care for her, including the researcher.  4.4 Casing the Case In accepting case studies as method, and casing as a methodological step (Ragin, 1992), the following section describes the process of casing the case study. In this study, cases were selected for their informational representativeness (Sandelowski, 2011). That is, I sought a combination of specific elements to be present to construct the case. In selecting the cases, I was not seeking to have diversity between cases; rather, I recruited cases with similar characteristics. Each case was cased along empirical, temporal, and spatial dimensions (Sandelowski, 2011). The first dimension defines the boundaries of the case inductively 72    through empirical research (Ragin, 1992) that describes the term advanced dementia as a constellation of specific cognitive and physical impairments. In addition, the term nearing the end of life was considered through the lens of advanced dementia that also delineated specific indicators of medical instability and frailty. While there is a fair bit of variability in the research literature in how the concepts of advanced dementia (also referred to as end-stage, severe, or very severe) and end of life are operationalized, the following criteria were used to case the case study along empirical lines.  For the purposes of this study, to be designated as a person with advanced dementia, the individual had to have a diagnosis of dementia as indicated on her medical chart and have been assessed as stage 7C – 7E on the Functional Assessment Staging Tool (FAST) and a six on the Cognitive Performance Scale (CPS).16 Both assessments had to reflect the resident’s current functioning (i.e., the assessments had to have been recently completed). In previous research, both the FAST and the CPS have been used to identify advanced dementia: The FAST stage 7C or higher is also used in the United States as hospice eligibility criteria (Porock et al., 2005) suggesting that this stage has some construct validity in identifying those nearing the end of life. The CPS is a measure of cognitive performance (not dementia per se) that is automatically generated from the MDS and has been used as inclusion criteria for a number of studies conducted by Mitchell and colleagues (Mitchell et al., 2004; Mitchell et al., 2006; Mitchell et al., 2007; Mitchell et al., 2009).                                                  16 The specific characteristics of advanced dementia and the problems with staging dementia have been described in Chapter 2: Review of the Literature  73    To be designated as nearing end of life, I specified that the person with advanced dementia would have two or more of the following clinical characteristics: swallowing problems; recent unintended weight loss; recurrent infections such as pneumonia or febrile episodes; recurrent or unhealed skin ulcers. These characteristics are associated with increased six-month mortality for those with advanced dementia (Mitchell et al., 2009; Porock et al., 2005).  In addition to being empirically cased, the case was also temporally cased. The manager identified potential resident participants by answering the “no surprise” question: “Is this person sick enough that it would be no surprise if he or she died within the coming year (or the coming few months)?” (Lynn, 2004, p. 44). Recommendations from palliative care experts in the United States (Weissman & Meier, 2011) and the United Kingdom (Aspinall et al., 2002) have suggested the surprise question both as an indicator to guide prognostication and for implementing palliative care to those living with a life-limiting illness. Residents for whom the answer was “yes” were recruited for the study. In addition, as I wanted to collect a sufficient data on the resident participant to facilitate a thick description of her experiences, my research protocol specified that if the resident died within three months of data collection, another resident would be recruited.  Finally, the case study was cased along spatial dimensions. Individuals with advanced dementia were recruited from a residential care facility. Ideally, the care facility would have a full complement of staff comprising nursing and allied health staff. However, I recognized that there might be some variation in terms of the configuration of staff mix. To that end, the exact mix remained unspecified. In addition, resident participants would also have a do-not-hospitalize advance directive, operationalized as a 74    two on the degree of intervention (DOI 2). A DOI 2 means that “therapeutic measures and medications to manage acute conditions within the limits of the facility” will be used (McGregor, Pare, Wong, Cox, & Brasher, 2010, p. 1160). The resident would not be transferred to a hospital for treatment unless the nursing home is not able to provide adequate comfort care, and the resident would not receive measures such as CPR or feeding tubes (McGregor et al., 2010). The decision for a DOI 2 was practically motivated as it was beyond the scope of this study to collect data in an acute care facility. 4.5 Entry into the field Two cases were recruited from a local residential care facility called the Manor.17 This site was known to me as I had worked with several of the Manor’s employees in other care facilities. However, it was recommended for this study by a member of the Manor’s board of directors who participated with me on an end-of-life community reference committee for our local health authority. When I described my study to her, she immediately recommended the Manor. “Oh, you have to do your study at our facility. The care there is fantastic!”  She gave me the name of the Manor’s executive director (ED); I contacted the facility in June 2013 and arranged a meeting with the ED and the director of care (DOC). During this meeting, I became increasingly cognizant that this facility would be an excellent fit with the goals of this research and was excited to learn that they were eager to be involved.  I also recruited another case from another local care facility. The Lodge was known as a leader in the field of person-centred care and was very involved with the                                                 17 Facility and individual names are pseudonyms. 75    research community. I arranged a meeting with facility management, one of whom I knew from my work with a local health authority. While I received considerable buy-in from the management, on a subsequent meeting to work out the details of resident recruitment, I was concerned that recruitment would be problematic due to the specificity of the case criteria. Recruitment did turn out to be problematic and the Lodge was dropped from the study as will be described in the following section.   4.6 Recruitment Processes and Sample 4.6.1 Recruiting Resident Participants Recruitment for the first case study from the Manor occurred in the following way. I first met with a manager who identified five residents who fit the criteria, all of whom resided on the special care unit (SCU). However, the manager suggested that I also consult with the staff who “knew the residents best” and a meeting was organized with the staff of the SCU. During this meeting, I explained the participant criteria and the staff, as a group, identified which residents they thought would be appropriate. There was a great deal of concurrence between the residents whom the staff named and those identified by the manager.  The manager suggested that rather than sending out the recruitment letter (see Appendix A) to the proxy decision-makers18 as per my study protocol, she would call each proxy and ask permission for me to call them. As such, I identified the order in                                                 18 In legal terms, resident participants were deemed to not have the capacity to consent to participation in this research (Hall, 2009). Thus, consent was sought from each resident’s proxy decision-maker (proxy). Proxies are legally sanctioned to make decisions on behalf of the resident. 76    which the proxy would be consulted—starting with Ella—the resident who staff thought would be best suited to the research. Margaret, Ella’s daughter and proxy decision-maker, agreed to a meeting. We met for 2.5 hours and Margaret was quite eager for Ella to be involved in the study and consented to her participation.  Recruitment for the second case began at the Lodge where facility management sent out recruitment letters to the proxies of potential residents after which the proxy contacted me directly. I was not aware of how many letters went out; however, I received three phone calls from interested proxies. Of these three, only one resident seemed to align with my criteria and I met the proxy face-to-face and enrolled his mother Molly in the study.19 However, during the time between recruitment and the commencement of participant observations, Molly had a significant health event and was sent to the acute care hospital (despite having a DOI 2). When she returned to the care facility, Molly’s plan of care shifted to palliative care. Some staff thought she would pull through as she had in the past, and so I commenced observations. Molly died within five days of her return to the facility where I had the honour of keeping vigil by her bedside and holding casual conversation with care staff over three of those five days. Because I had gathered very limited data on Molly, I began recruitment processes over again. However, in two months I received only one call from a potential participant’s proxy. I recruited this resident but upon visiting with her for the first time, I realised that the resident did not have advanced dementia, though she did have the indicators of nearing the end of life. I un-enrolled this resident explaining to the proxy and facility management that this participant, while an interesting case, was not suitable                                                 19 Molly was the subject of the Prelude opening this dissertation. 77    for this study’s purpose. I became increasingly concerned that I was not going to recruit an appropriate case study from this site in a reasonable time frame. In discussing this dilemma with my committee it was decided that I drop this site and recruit another participant from the Manor. This did not affect the goals for the study, as I was not intending to make between-site comparisons.  The second case from the Manor was recruited four months after the study started. Following an analytical line of nonverbal communication and embodied expressions, I asked the manager to identify someone who displayed different expressions compared to Ella, the first case. The intent was not to make cross-case comparisons, but rather to explore the notion of nonverbal communication more fully. As such this case was empirically cased, in addition to participant criteria as described earlier. The manager identified Elizabeth, who resided on the complex care unit (CCU), and contacted her husband Thomas, who was her proxy decision-maker. Thomas visited her three times a week and we arranged to meet on one of his visitation days. He was enthusiastic about Elizabeth’s involvement in the study and consented to her participation.  4.6.2 Recruiting Family Participants To understand how family constructed the resident’s experiences, I recruited family members who most frequently visited the resident participants. In both cases, this was only the proxy. I discussed their participation in the study (interviews and participant and video observations) at the same time as I recruited the residents, and both consented20 to participate, but not to video observations. I developed a close working relationship                                                 20 See Appendix B for written consent form. 78    with both proxies, conversing with them frequently, whether face-to-face or on the phone. Often these conversations were not oriented towards formal data collection (i.e., interviews), but to maintain rapport and a trusting relationship. In some ways, I became their companion as well as a companion to their family members. The evolution of this relationship will be explored in the section 4.9.5 Negotiating Roles and Relationships. 4.6.3 Recruiting Staff Participants Once I recruited the case, I purposively selected carers21 to participate in participant observations, video observations and interviews. The carers who most frequently interacted with the two resident participants were recruited as key informants, as they were most closely connected to the day-to-day care of Ella and Elizabeth. These carers were selected from all shifts and across the nursing and allied health program. I initially planned to meet with the carers as a group to discuss their participation in the study; however, to maintain some degree of privacy about who was participating, I opted to meet with carers individually. In total, 26 carers consented22 to participate in the research, 17 of whom consented to video observations (though not all were filmed), and three who consented only to traditional observations. Table 1 outlines select demographic data of the carer sample. Care aides constituted 58% of the sample; the majority of the carers self-identified as Asian/South Asian ancestry; self-identified their spiritual background as Hindi, Sikh or Buddhist; and were between the ages of 41-50. Half the                                                 21 I use “carers” as a generic term to describe all staff participants. In my analysis, to protect participants’ privacy I delineated between three job categories of carers: care aides (CA), nursing and management staff (NS), and allied health staff (AH). All carer participants were assigned a code (rather than a pseudonym) according to their job category.  22 See Appendix B for written consent form. 79    carers had worked at the facility for 10 or more years (average = 13). These demographics reflected the larger group of carers who worked at the Manor.  Table 1: Carer Sample Characteristics Carer Sample (n=26)  n % Job Category     Care aides 15 58  Allied health 6 23  Nursing/Management 5 19 Cultural Background     Asian, South Asian 22 85  Caucasian 4 15 Spiritual Background     Hindi, Sikh, Buddhist 12 46  Judeo-Christian 6 23  Not identified 8 31 Age Range (years)     20-30 3 12  31-40 2 10  41-50 12 43  51-60 9 35 Job Tenure (years) 0-5 9 35  6-10 4 15  10+ 13 50  Attrition of key informants was minimal during the 18-month data collection. One informant, who consented to participant observation/casual conversation only, left the Manor before the study was completed. Her data remains in this dissertation. It is important to note that only three men were recruited. This was reflective of the demographics of the care staff overall. To protect their identities, I refer to the men as “she” as per the request of one of the male staff participants.  As I recruited carers who most frequently interacted with Ella and Elizabeth, the sample is distributed between them unevenly. This was a function of how the 80    organization of care differed between the units where the two residents resided. Because the SCU care aides’ workload was scheduled on a rotating basis, all SCU care aides worked with Ella. As such, all regular and part-time care aides and LPNs (n=11) were invited to participate in the study. All staff consented to participate in the study, though two consented to observations only. By contrast, on the CCU where Elizabeth resided, only four out of seven care aides consistently worked with her and all four consented to participate in this research. An additional three care aides were recruited, two of whom worked with Elizabeth on a less consistent basis, but had known her since she moved into the Manor, and one casual care aide who worked with both Ella and Elizabeth. As I did not observe a great deal of RN and LPN interaction with Elizabeth (other than the occasional medication assist), I sought verbal consent for their participation. Likewise, I sought verbal consent for participant observation from visitors as well as other staff who interacted with the residents on an irregular basis (e.g., casuals).  All nursing managers consented to participate (n=2). Six allied health carers consented to participate. This group consisted of four out of eight regular allied health staff, one out of five contracted staff, and one entertainer.23 Allied health participants reflected the diverse range of allied health services at the Manor: recreation therapy, music therapy, pastoral care, dietary and food services, and rehabilitation therapy. Of the allied health carers who did not participate, none were observed to provide services to Ella and Elizabeth; thus they were not invited to participate in the study.                                                  23 To conceal her identity, from here on the entertainer will not be differentiated from the rest of the allied health group. 81    4.7 Data Collection Procedures Ethnographic case studies typically employ multiple techniques to gather a rich array of data to facilitate a robust understanding of the complexity of the case (Stake, 1995). The data collection methods in this study comprised: participant observations and informal conversations; video observations and reflexive video feedback interviews; formal interviews; and the review of documentation related to care. This data collection strategy elicited a wealth of data and helped contribute to the trustworthiness of the findings through triangulation of methods.  4.7.1 Participant Observation Participant observation consists of on-site research during which the researcher is present in the setting to observe naturally-occurring activity and interaction to develop insight about the phenomenon under study. While the definition of participant observation is difficult to pin down (Atkinson & Hammersley, 1994), Junker (1960) outlines a fourfold continuum: complete observer, observer as participant, participant observer, and complete participant. This continuum points to several dimensions of ethnographic observations: first, the extent to which the researcher and the research is known to those who are being studied (as well as nonparticipants); second, the sorts of activities the researcher is engaged in whilst in the field; finally, the degree to which the researcher adopts an insider-outsider perspective (Atkinson & Hammersley, 1994; Junker, 1960; Keith, 1986). In this study, residents, family, and care staff were apprised of my role as I wore a nametag with the words PhD student and researcher. Many nonparticipants asked about my study and I was transparent about what I was doing and 82    the reasons behind the study. While many nonparticipants were interested in who was participating in the study, I did not divulge who was involved. In accordance with Junker’s (1960) continuum, I primarily adopted a participant observer role, though throughout the data collection process, I found myself at different points along the continuum, even at times being a complete participant. Generally, the shift from observer to complete participant paralleled the shift from outsider to insider position within the environment. As Keith (1986) noted, the shift from outsider to insider is characterized by the movement from a macro view of the environment that gives foundation to the research to a more focused inquiry through the use of participant observation and interviews. These prepare the researcher for a deeper engagement with the case itself necessary for the analysis and the ethnographic description.  My work and previous research experiences24 in residential care somewhat expedited the process of moving from an outsider to an insider perspective. That is, while I was an outsider to this specific site, I was very familiar with the governance of residential care, the organizational environment, the staff-mix that I would encounter, and the social programming that would occupy a resident’s day. Thus, while I entered the place as a stranger, the space was very familiar. During the initial time of engagement in the field, this grounding was integral as it contributed to developing rapport and trusting relationships with care staff and family. As the study progressed, and considering the prolonged immersion in the field, I became more intimately connected and familiar with                                                 24 I was a Music Therapist and worked in nursing home environments for about 15 years. My Masters research was an ethnographic study conducted in two nursing home settings. 83    the site and I became more of an insider in the traditional sense of that term (Borbasi, Jackson, & Wilkes, 2005). Moreover, I became more of a complete participant.  During the fieldwork period, I participated in the daily routines of the Manor. On any day, I helped serve meals, helped make Ella or Elizabeth’s bed after morning care, assisted Ella or Elizabeth to activities, participated with them in activities, and conversed with carers, family members, and residents (including nonparticipants). On several occasions, and after instruction from care staff, I assisted Ella with a thickened beverage or Elizabeth with afternoon tea. I also helped care staff with transferring Ella or Elizabeth (also with instruction). Concurrently, I observed the full spectrum of activity: morning and evening care routines; recreation and spiritual care activities; meal times; and times when the resident was alone in her room or not in activity. I partially observed toileting and bathing routines, though I positioned myself discreetly so that I did not observe the resident in complete undress. Similarly, during morning and evening routines, I would turn my back when the resident was undressed. Participant observation took place on all days of the week in order to maximize variation in context. While I anticipated observing across all shifts, because Elizabeth and Ella were assisted to bed following dinner (around 6:30 p.m.) and did not have troubles sleeping once in bed, I primarily observed during the day and early evening. Over the first nine months of data collection, I adopted a fairly intensive observation schedule, observing up to six hours per day up to six days a week. After nine months at this intensity, I pared back the observations to up to four hours per day three to five days per week. As I became more intently involved in data analysis, I pared back the observations even further, going into the facility once or twice a week for two to four 84    hours, collecting data that would follow specific analytical lines and data that would contribute to the resident participant’s overall temporal narrative. This culminated into well over 500 hours of ethnographic observations. I observed Elizabeth for 210 hours and Ella for 427 hours. The large discrepancy between the two is because I only observed Ella for the first three months of the research before Elizabeth was recruited. I was in the field for just over 18 months. The decision to stop data collection was motivated by a practical decision considering the amount of cumulated hours of participant observation; as well I believed I was not collecting any new data. The new data would come if the person moved into an active dying phase and I requested that, in addition to weekly check-ins, the facility management keep me abreast of the residents’ health status.  4.7.1.1 Observational Field Notes as a Source of Data  Field notes constituted the primary source of data for this study. I made extensive jottings in the field recording the resident participants’ everyday activity. I documented their vocal expressions, body movement and facial expressions. I musically scored the rhythmic and tonal patterns of Ella’s vocalizations, and attempted to do the same with Elizabeth’s mumblings. During care routines, I recorded what occurred and took notes on the coordination of action and interaction between carers and the residents. I took note of the sensory environment, particularly of the sounds that permeated the space which Ella or Elizabeth resided. I also recorded discursive data and paid a great deal of attention to what was said, who said it and to whom, and how it was said. Finally, by way of reflexive jottings, I took note of my emotional and embodied responses to what was occurring in the field.  85    While my jottings in the field were fairly developed (as there were periods of considerable quiet in both Ella’s and Elizabeth’s lives), the task of rendering them into accurate and fully realised field notes was somewhat daunting. One suggestion in the literature is to develop the field notes immediately following the observation period (Emerson et al., 2011). In this study, as I often spent all day at the facility, I took discrete breaks throughout the day to expand upon the jottings. Other times, notes were developed when I returned to my office. The fully developed fieldnotes were vivid sketches of the flow of Ella’s and Elizabeth’s daily life, a narrative constructed around the sequence of interactions, activity, and experiences throughout their day. Though I believe I was able to textually recreate what I observed and heard with much accuracy, I concede that, as Emerson et al., (2011) stress, all fieldnotes are reconstructions that transform witnessed events into researcher-generated text, and even spates of talk, jotted as close to the speaker’s words as possible, are reconstructed from a partialized memory (Reissman, 2008).  4.7.1.2 Casual Conversation as a Source of Data  Conversations that were not pre-planned and occurred in-situ were a significant part of the participant observation method and were documented in my fieldnotes. In this study, casual conversation involved clarification and commentary on events experienced during the observations. In particular, I asked carers questions about the interactions that they had with Ella and Elizabeth, shortly after the interaction occurred. I asked these questions especially if inconsistencies were observed, for example, when practice did not align with their beliefs about care or align with what the Manor’s expectations were. In addition, casual conversations were initiated by care staff who were eager to tell me 86    about their experiences with resident participants that I may not have observed. Care staff also initiated conversation about their experiences working in this particular care environment. Finally, care staff freely offered their perspective about end-of-life care issues. For example, a care staff asked me to read an article about the B.C Supreme Court Decision on the Margot Bentley case,25 and offered her perspective on the ruling. Considering the co-constructed and dialogical nature of this research, these naturally-occurring exchanges enabled emergent meanings to flourish and offered essential details that supported the aims of this study.  4.7.2 Interviews Ethnographic interviews allowed for exploration of specific issues surrounding care interactions with Ella and Elizabeth and how care staff and family constructed the resident participant’s experiences. In addition, interviews explored care practices and perceptions about end of life, the meaning of suffering, and socio-cultural perceptions of dying and death. These data not only brought much-needed context to participant observations but also illuminated the myriad of socio-cultural factors related to dementia, dying, and death.                                                  25 Margot Bentley is a woman living with advanced dementia in a BC care facility. Many years prior to her admission, Mrs. Bentley wrote a living will that explicated that if there was “no reasonable expectation of her recovery from extreme physical or mental disability” all nourishment and liquids were to be withheld (Bentley v. Maplewood Seniors Care Society, 2014 BCSC 165, para. 5). As the carers at the facility continued to spoon-feed Mrs. Bentley a pureed diet, the family took the facility to court to enforce what they believed to be clearly stated wishes. The BC Supreme Court ruled in favour of the facility organization stating that withdrawing nutrition and hydration would “constitute neglect within the meaning of the Adult Guardianship Act” (Bentley, 2014 BCSC 165, para. 153). The family appealed the decision; however, the Court of Appeals ruled in favour of the facility organization (Bentley v. Maplewood Seniors Care Society, 2015 BCCA 91). 87    While casual conversations provided spontaneous opportunities for data gathering, formal interviews provided a purposeful and structured space to explore informant perspectives more deeply. In designing the interviews and considering when they would occur in the research process, I was committed to the idea of creating a dialogical space not just to gather information, but also to develop a space for interaction. Thus, interviews did not take place immediately upon entering the field. Rather, I felt it not only necessary to develop a better understanding of Ella and Elizabeth prior to interview conversations, but also necessary to establish a relationship with informants first. Though interviews are not “conversations between equal partners because the researcher defines and controls the situation” (Kvale, 1996, p.6), I sought to position myself and the informants as collaborators in the generation of knowledge (Holstein & Gubrium, 2008).  I developed an interview guide constructed on emergent issues that arose from the participant observations and document review as well as issues drawn from the literature (see Appendix C). However, I found that the most successful interviews were more conversational in style, adapting my questions to elicit the informants’ experiences from their perspectives rather than to force upon them preconceived categories. The emphasis in the interviews was to illuminate the experiences of the informant in relation to Ella or Elizabeth, and how the informant constructed and ascribed meaning to the resident’s experiences. Moreover, I was interested in how knowledge was co-constructed through the interview process. 88    4.7.2.1 Participant Interviews  Staff who were most involved in the care of Ella and Elizabeth were invited to participate in the interviews. Staff included frontline nursing and care aide staff, allied health members, and managers. Twenty-three carers consented to in-depth interviews including: 12 full-time care aides, one casual care aide, three LPNs, two management staff, and five allied health staff.  I also invited Ella’s and Elizabeth’s family members to participate in interviews. As I was interested in how family constructed the resident’s experience, I only invited those who visited on a regular basis or were involved in care planning. Thus, only the two proxies participated in the interviews as other family were not involved. 4.7.2.2 Structure of the Interviews Informants were asked to participate in up to two interviews. The first interview with carers was constructed as the primary interview to explore how they understood the resident’s experiences as well as their perspective on issues related to end-of-life care. The second interview was temporally designed to coincide with when the resident participant appeared to be moving towards the last months/days of her life (i.e., when carers were thinking that the resident’s condition was deteriorating). The focus of this interview was to explore emergent meanings associated with the death of this particular resident. All the carers involved in Ella’s care took part in the two interviews while those involved with Elizabeth did not, reflecting key differences in Ella’s and Elizabeth’s 89    health status over time.26 The interviews were audio-recorded and lasted from between 30 to 120 minutes (average 40 minutes). A total of 36 carer interviews were held.  Carers were given the choice to be interviewed on- or off-site, and all but one chose to be interviewed on-site. Two challenges emerged by holding on-site interviews. First, interviews were a challenge to schedule because it was difficult to predict when a carer would have time in her day to have a conversation without interruption. Interviews took place during the carer’s breaks, when residents were asleep, or when residents were engaged in other programming. I was fully aware that I was cutting into the carer’s work time, though I received support from management to do so. What was very significant, and quite typical, was how the carers accommodated each other to be involved in the interviews. This was not unlike what I observed on any given day; carers supported each other by reorganizing their tasks or temporarily taking on a heavier workload. Second, it was difficult to find a space that was free from distraction as well as a space that maintained privacy and confidentiality. I was especially concerned with holding interviews on the respective units, as management were often present on the unit. While I let staff direct me to the spaces that best suited them, this posed methodological and ethical dilemmas. On the CCU, some nursing staff opted to be interviewed in a charting space located at a work desk in the corner of the resident bathing area (i.e., the location of the tub, shower, and several toilets). On one occasion, much to my surprise, a resident was in the space (concealed by a privacy curtain). When I realised that the resident was in the space, I stopped the interview and suggested that we find a different                                                 26 While Elizabeth’s health status did deteriorate, carers did not construct this to be indicative of the dying process. She did experience a significant health event after data collection completed. After consultation with my committee, I decided that additional data would not significantly shift the predominant themes in the rendering of her case.  90    location. The carer was reluctant to be interviewed elsewhere. “It’s fine,” she said, “he can’t hear us…we always chat in here.”  While the staff participant did not see the inappropriateness of the situation, I suggested that we hold the interview at a different time. In reflection and dialogue with my counsellor,27 I was sensitized to the idea that this particular space was the domain of the carers; i.e., this is where she was perhaps most comfortable and where she had control and power. By insisting on holding the interview in this space, she was asserting control of the interview situation and perhaps levelling the power dynamics inherent in interviews. From this situation, I became alerted to the methodological (and ethical) significance of place in terms of how geography and social dimensions of space can potentially shape meaning making and how knowledge is produced (Anderson, Adey, & Bevan, 2010; Ingold 2011). This emerged most poignantly in the interview situation, but also in participant observation and video observations, where place and being emplaced as a researcher arose as being significant. This will become more evident to the reader in the subsequent findings chapters.  While the interview process progressed in a typical question-answer and back-and-forth manner (Kvale, 1996), in two interviews I became increasingly cognizant of the degree to which I participated in the dialogue, to the point where I wondered “whose voice is this really” and to what degree was I “putting words in someone’s mouth?” In part, this arose because for some informants, articulating their perspective was very difficult. As they struggled to find the words to express their thoughts or if they only                                                 27 I sought the support of a life coach as part of a reflexive approach to research practice. See page 113. 91    supplied one-word answers, I used prompts, leading questions, and probes to help them communicate their thoughts. However, often these were taken up as expressions of their reality in the sense that the informant would agree with my question (or statement) rather than offering more information. I realised that my interjections arose because I felt uncomfortable with what I perceived to be a great deal of embarrassment on the part of the informant when she would stumble over words. By my interjections I was hoping to make the informant more at ease. In the subsequent analysis of interview data, I cultivated a deep appreciation for the interactional and constructed nature of this research. As Kvale suggests, “The task is…to recognize the primacy of the [leading] question and attempt to make the orienting questions explicit, thereby providing the reader with the possibility of evaluating their influence on the research findings and of assessing the validity of the findings” (p.159). I also conducted interviews with each resident’s family member. I interviewed Ella’s daughter twice, each lasting between 2-2.5 hours. As Elizabeth’s husband was a steady and frequent visitor (visiting 3 times per week) and we had many lengthy conversations during these visits, I only formally interviewed him once, which lasted for 90 minutes. The interviews started from the beginning of the dementia process, highlighted key transitions related to the progression of dementia, and explored how they made sense of Ella or Elizabeth’s present situation, and the looming prospect of the person’s death. The interviews with family members were not meant to be therapeutic, but on many occasions they expressed to me the value (and relief) of having these conversations.  92    In asking sensitive questions about dying and death and informants’ personal journey with Ella and Elizabeth, I was well aware that strong emotions might be elicited. In addition to being empathetic, employing good listening skill and validating their feelings, I gave informants the option of stopping the interview (which none did). I also gave interested informants a list of grief and bereavement resources available in their local jurisdiction and I had made prior arrangements with a therapist colleague for support service for staff if needed. One care staff took the printed resource, though when I offered the name of the therapist, she did not feel that was necessary. In spite of addressing sensitive topics, the majority of care staff and both family members reflected back how much they appreciated and were grateful for the opportunity to talk about their experiences.  4.7.3 Video Observation Despite the burgeoning use of video observations in health research, video data appears to have only made smalls steps to inform scholarly research into care practices within, and experiences of, dementia care in nursing home environments. Examining the research landscape, video observations have been used in several ways. First, video data is used to understand verbal and nonverbal expressions of challenging behaviours (Beck et al., 2011) and nonverbal expressions of pain (Husebo et al., 2007). Second, it is used to evaluate the effectiveness of interventions such as music (Ragneskog et al., 1996, Norberg, Melin, & Asplund, 1986) and theatre (van Dijk, van Weert, & Dröes, 2012) as well as to evaluate care practices such as bathing routines (Skovdahl, Sörlie, & Kihlgren, 2003) or personal care routines (Engström & Hammar, 2012; Hammar, Emami, 93    Engström, & Gotell, 2011). Finally, it is used to examine social interaction patterns such as staff-resident communication (Asplund, Jansson, & Norberg, 1995; Carpiac-Claver & Levy-Storms, 2007). Taken together, this body of research has shown video data to be instrumental for reaching a deeper understanding of the relational and individual experiences of persons with dementia who reside in nursing homes. However, the researchers rarely fully explain their methodological approach, in-situ practices, and ethical concerns related to the use of filming in a nursing home setting. As such, what follows is a discussion of the importance of the visual, and the process I used in using a video camera, in this research. The use of video in ethnographic research offers a route to understand observable phenomena in new and evocative ways. In her seminal work, Doing Visual Ethnography, Sarah Pink (2013) wrote, “visual ethnography cannot be defined as a simply observational method” (p. 39) as this invokes a detached observer and an objectifying method. Rather, it is through the researcher’s direct and embodied engagement with the research environment that visual images are produced. That is, they are composed from the position of the person holding the camera (MacDougall, 2005) as she participates in the social, sensory, and material environment in which she is emplaced (Pink, 2013).   I was motivated to use video for several reasons: to help me better understand the embodied expressions and experiences of Ella and Elizabeth; to understand staff’s embodied knowledge of care work; and to explore the sensory environment and the ways it may impact a person’s experience. I also wanted, in some way, to accentuate the voices of Ella and Elizabeth and a way for the reader/viewer to hear their voices. Participant perspectives are typically accentuated through the use of verbatim quotes; this was not 94    possible in Ella’s and Elizabeth’s case. However, I was reluctant to use video, and the notion of objectifying the resident was a foremost concern. While this was not limited to video, as I often felt that I risked objectifying others in participant observations, focusing the camera’s gaze on a specific resident participant made me consider whether video was indeed the right tool for this research. I weighed the burden/benefits to the participants and concluded that by taking a collaborative and reflexive stance the problem of objectification could be partially overcome. Moreover, by becoming a participant in the resident’s social world (as described earlier), I opened the possibility for doing research with, instead of on, the person (Dewing, 2007). 4.7.3.1 Video Footage as a Source of Data Video observations were used judiciously to follow resident activity over time and in different contexts, to illustrate specific analytical lines, and to highlight a particular narrative. To that end, video observations were conducted when the residents were alone, in social spaces or in personal care. As will be discussed in section 4.9.4 Consent Processes, what was filmed was negotiated between the proxy, carers, and myself, as well as through ongoing assent with resident participants. In addition to negotiated consent, to develop a better understanding of what would be appropriate to film in consideration of the specific resident, I also considered biographical information, such as, for example, how the person might have responded to a camera (video or otherwise) in the past, and how the person responded to nursing home care over time. Building such understanding of the person was integral to this research to ensure that the person’s values and preferences were considered. Finally, my own moral radar guided me to determine whether it was right to film in particular situations as they emerged. 95    The video itself was not taken in a systematically controlled way (i.e., all video was not taken in the exact same way) as limits in the physical and social environment pressed upon the act of filming itself. For example, in filming a meal interaction, in addition to ensuring that I was not obtrusive, the way I framed the interaction was dependent on how non-participants moved through space. To ensure that others were not inadvertently filmed, at times I zoomed in on the resident participant’s face, or on her food, or I concealed the lens until the nonparticipant moved. To keep track of contextual elements, and methodological or ethical issues that arose whilst filming, I kept fieldnotes following video observations.  I took about 30 hours of footage with just under 6 hours of usable video footage of Elizabeth and 13 hours of usable footage of Ella. The discrepancy in the amount of time was due, in part, to the challenge of filming Elizabeth in the social spaces of the Manor. It was very difficult to ensure that nonparticipants were not in the frame. With Ella, this was considerably easier to control as there were fewer people in her space and she was often seated alone. 4.7.3.2 Equipment I used a small user-friendly hand-held video camera that recorded digital footage (rather than taped footage). I also used two types of tripods – one that could be quickly assembled on a tabletop or other flat surfaces, and another flexible tripod that could bend and rotate such that the camera could be positioned on a wide range of surfaces, including bed rails and chair arms. All digital files were uploaded to a password-protected computer and external hard drive after I returned to my home office from the Manor; 96    subsequently, files on the camera’s memory card were immediately erased. I catalogued and time-coded the footage for ease of retrieval and editing.  4.7.3.3 Video Reflexivity as a Source of Data Video reflexivity is a process whereby video footage was played back to participant carers shown in the footage and family members, in their role as the resident’s proxy decision makers, for review and discussion. By inviting participants to become involved in the analytical process, I made visible my own observations and positioning and created a powerful channel for sharing of knowledge (Pink 2013). Thus, critical to the analysis of video data was an understanding that my conclusions were not final (Frank, 2012), but open to emerging perspectives; and subsequently, reflexive sessions produced another layer of knowledge. This crossing of boundaries (the researched as knowledge producers) created further opportunities for co-construction of meaning (Carroll, 2009).  Reviewing footage with participating carers and family members served two purposes: first, to offer the opportunity for participants to comment and reflect upon what was revealed on screen; and second, to solicit consent to share footage with others. Reviewing minimally-edited video footage with participants provided them with the opportunity to discuss, share, analyse, and even problem-solve daily practice. I held one reflexive session with each carer who was part of video observations. I held two sessions with Elizabeth’s husband, and three with Ella’s daughter. In terms of carer participants, I asked them to reflect on their practice in order to facilitate a greater understanding of how they interpreted and responded to the residents’ nonverbal cues, and of how they worked together (in the footage of two-person assists). 97    What staff said about themselves and their practice was very revealing, not only of how they perceived themselves in practice and their relationships, but also of how practice was embedded in the larger context, including facility policies and the physical environment, which impacts the care encounter. For example, in discussing footage of meal time assistance, a nursing staff participant commented that while she was very pleased with how she was interacting with Ella, she thought that, based on her understanding of the infection control policy, she might be sitting too close to Ella. I would not have known this detail of the policy without having had this conversation with the care staff participant. Accordingly, I was concerned that by sharing this video footage with a wider audience, she would be negatively impacted. In raising the issue with this participant, she relayed that she was not concerned; she was proud of the work that was depicted in the footage and consented to share that footage.  Likewise, sharing video with family members was also a powerful feedback method for revealing their understanding of their relatives’ experiences thereby giving new meaning and interpretations that would otherwise remain untapped. However, with both staff and family members, the sharing experience requires a great deal of sensitivity on the part of the researcher. Decisions about what footage to share and when to share it needed to be carefully contemplated in advance, particularly with footage that may evoke a strong emotional reaction. For example, some footage of Ella showed her to be in discomfort or pain. In preparing for a reflexive session with Ella’s daughter, I alerted her to this and asked if she would like to see this footage. She did want to view the footage, as this was “part of Ella’s reality” [FN: 16.09.14]. However, it was not until the final 98    reflexive session that she seemed open to viewing it. The following fieldnote illustrates what transpired: During the viewing of the video, I was alerted to Margaret’s body language, which seemed to be telling me that she was upset. I stopped the video, and asked if she was okay, she reassured me that she was, but asked if I could turn the volume off. I did so, and she visibly relaxed. After the video, she commented, “it’s really obvious that mom’s in pain…you can see it in her face, the grimaces, the shifting of her body…” I replied, “Yes, even with the sound off, it seems pretty clear.”  Margaret relayed that from her visits, she knew that her mom was in pain, but the video made it more immediate. In the moments of visiting her mom, she was more concerned with providing comfort, than paying attention to the facial expressions. [FN: 04.03.15]  With the goal of making visible the lives of study participants to a wider audience, integrating the visual (either as a moving image or a still photograph) into this dissertation was given much consideration. I was concerned that in re-presenting participants, I was inadvertently perpetuating negative stereotypes of people with advanced dementia and the dementia care environment. In seeking consent from the proxies and carers this risk was discussed and a fruitful discussion about representation and authorial voice resulted. It is often assumed by researchers that privacy is of the utmost importance to participants and proxies when in fact other values may be equally or more significant (Bombard, Cox, & Semaka, 2011). This was indeed the case: In addition to being very selective about what footage could be shared (if at all), I offered proxies and carers the possibility of heavily editing video (e.g., blurring faces). Several staff and a proxy [e.g., FN: 06.02.15] remarked that this would defeat the purpose of making visible the experiences of those in the video footage, because embodied expression, especially facial expressions, were so integral to communication and understanding experiences. 99    4.7.4 Document Review To explore how Ella’s and Elizabeth’s experiences were constructed in written text, I reviewed several documents related to their respective care: 1) daily progress notes; 2) quarterly assessments; 3) care plans; 4) physician notes; 5) clinical indicators such as vital signs and medication reconciliation; and 6) documentation leading up to the resident’s admission such as psychiatric consultation referrals and home care assessments.  In addition to clinical documentation directly related to Ella’s and Elizabeth’s care, to understand the question of how end-of-life care is constructed at the Manor, I also reviewed specific policy/procedural documents including: the local Health Authority’s End-of-life Care Manual; the Manor’s palliative care manual for staff; and the manual for family. In addition, I kept note of the documentation fixed to walls, notice boards, and elevator doors. For example, I explored the documentation related to ethics and religious perspectives on end-of-life care that was posted on staff notice-boards. I read the resident activity calendars posted on the walls, announcements of deaths of residents, and artwork that depicted integrating the Manor’s Mission into everyday practice. Finally, I noted the inspirational quotes posted in the elevators, some of which carers quoted or referenced in their interviews or in casual conversation.  4.8 Analytical Procedures 4.8.1 Case-Oriented Analysis  To maintain intimacy with the case, I conducted a case-oriented analysis rather than cross-case analysis. Thus, my attention was fixed on the unique assemblage of each 100    case as a whole rather than combining them conceptually (Ragin, 2000; Sandelowski, 2011). I was motivated to do