Understanding Family Responses to Life-Limiting Illness: In-Depth Interviews With Hospice Patients and Their Family Members

Abstract / Understanding family dynamics is a key component in providing comprehensive care for persons with progressive illnesses and their caregivers. The purpose of this study was to investigate what families experience during an advancing illness and to describe their patterns of response. In-depth interviews (n=108) were conducted with families two weeks after hospice admission. Interviews were tape recorded, transcribed, and analyzed using qualitative methods. Six modes were distilled: reactive (illness generates intense responses), advocacy (vulnerability ignites assertive actions), fused (illness and decline are shared experiences), dissonant (diametrically opposed viewpoints cause struggle), resigned (decline and death are anticipated), and closed (outward responses are impassive). Three events triggered movement from one mode to another: (1) functional changes, (2) crisis events, and (3) provider communication. Providers who understand varied family reactions can ease the patient’s suffering, assist relatives in providing effective care, and prepare them for the approaching death.

Rsum / La connaissance de la dynamique familiale demeure l’un des principaux lments dont il faut tenir compte lorsqu’il s’agit de dispenser une gamme complte de soins aux personnes en phase progressive de la maladie et leur famille. Le but de cette tude tait de connatre le vcu des familles durant cette phase et de dcrire les modles de comportement auxquels elles ont eu recours pour y faire face. Pour y parvenir nous avons fait des interviews en profondeur auprs de 108 familles, deux semaines aprs l’admission du malade en soins palliatifs. Les rencontres ont t enregistres, transcrites et analyses selon la mthode qualitative. Six types ou modles de comportement ont t relevs : la raction (la maladie gnre des ractions intenses), la dfense (la vulnrabilit soulve des actions affirmatives), le fusionnement (la maladie et son dclin deviennent une exprience partage), la dissonance (les points de vue diamtralement opposs causent des msententes), la rsignation (le dclin et la mort son apprhends) et le repli sur soi (les ractions extrieures sont de nature imperturbable). Trois vnements marquent la dynamique familiale et font que l’on passe d’un modle de comportement un autre soit : les changements fonctionnels, les vnements dramatiques, et l’information transmise par le personnel soignant. Lorsque ce dernier saisit bien la nature de ces diverses ractions familiales, il peut allger la souffrance du patient, assister les proches dans la dispensation de soins efficaces et les aider faire face la mort imminente.

INTRODUCTION

Understanding and responding to family dynamics is a key component in the provision of comprehensive care for persons with progressive illnesses (1,2). Awareness of the possible emotional and behavioural reactions which accompany the turning points of an advancing illness can assist practitioners and the families they encounter. For providers who must approach families largely as strangers, knowledge of family dynamics can ease the challenge of communicating bad news such as a poor prognosis or limited treatment options. Emotional, angry, or resistant reactions, which may short- circuit communication and be labelled “difficult” or “dysfunctional”, can then be understood as a reflection of fear, uncertainty, or anxiety (3,4). The ability to communicate about a loved one’s terminal state has been found to emerge differently in different family members, primarily because they first develop awareness as individuals, then together as a family (5,6). Families benefit from interactions with providers who understand that individual members’ comprehension may be uneven, and who can help ease the difficult transitions that accompany decline and loss (7,8).

The Palliative Care Task Force has published standards (9) which highlight the importance of providing care for patients and their families as a unit. Moving in this direction, researchers have explored families’ perceptions of good end-of-life care, documented situation-specific stressors, and worked to develop an accurate way of assessing caregivers’ quality of life in the final stages of an illness (10-12). Previous family research in palliative care identified behaviours that may signal risk for complicated bereavement reactions (13). The nature of research in end-of-life and palliative care focuses solely on the important final stage of the illness, but this approach diminishes attention to the cumulative effects of prior illness-related events. The naturally occurring variability of families’ responses to the turning points of a life-limiting illness has been largely unexplored. By learning from the experiences of people who were experiencing a life- limiting illness, the present study moved toward a conceptualization of how the individuals in a family respond together. Following a brief literature review of families and life-limiting illness, this paper presents a typology of responses that emerged from the analysis of interviews with hospice patients and their relatives.

LITERATURE REVIEW

A life-limiting illness is diagnosed in an individual, but its effects reverberate through the family system (2,14,15). Illness- related changes can profoundly affect family dynamics, beginning at the onset and progression of symptoms, intensifying with the need for care provision, and continuing through both the final stages of advanced disease and bereavement (16,17). Progressive illnesses such as cancer have been conceptualized as a series of stages in which families experience situation-specific crises (18,19). Events which precipitate these crises can include: diagnosis, treatment (e.g., chemotherapy or radiation) and its endpoints, remission, recurrence, and the terminal decline (16,20-22). Other unforeseen illness- related changes that have been found to generate distress include: hospitalization, palliative surgery, and uncontrolled symptoms (pain, nausea, anxiety, or delirium) (23,24). These events heighten families’ vulnerability and distress.

Serious and progressive illnesses shift the focus of close primary and sometimes intimate relationships from shared tasks to a central focus on caregiving for the person with increasing dependence (25,26). Extensive research on caregiving in chronic illness indicates that these concerns become intensified when families provide care at life’s end (10,27-29). However, family responses to an advancing illness encompass much more than caregiving. While providing increasingly complex care, families also experience significant transitions and role losses as well as emotional preparation for the approaching death.

Patient-centred and family-focused palliative care has been advanced by Teno et al. (11) as a mechanism for easing the distress of an approaching death. Consideration of the family as a unit is the underpinning of palliative care, and each of the principles is also relevant for helping the patient-family unit throughout the stages of a life-limiting illness (30). Guided by the family systems perspective, this study conceptualizes the patient and family as a unit (31,32). The purpose of the study was to investigate what families experience during an advancing illness and to describe patterns of responses. This line of inquiry has implications for enhancing health care providers’ understanding of patient-family needs and reactions they encounter when treating seriously ill persons.

METHODS

Design

This exploratory, descriptive study employed qualitative methods to acquire an in-depth description of participants’ illness-related experiences in their own words. Previous research about family members of terminally ill people has often been conducted retrospectively using after-death interviews (see Addington-Hall and McPherson, 2001, for a review) (33). However, because perspectives about events change after death, this study was framed within the time period after treatment had ended and before the final stage of life had begun.

A systematic inductive analysis of family accounts of the stages of an illness was used to identify emergent themes and to develop a descriptive typology. Two research questions guided the study: What do families experience during the diagnostic and treatment phases of advancing illness? How do families respond to the events of a life- limiting illness?

Exploring the perspectives of the ill person and their relatives together is particularly important to advance understanding of the family dynamics. Two definitions are important. First, “life- limiting illness” was defined by hospice eligibility. Second, “family” was defined as either relatives or people in close or intimate relationships. The process of gathering different family members’ accounts made it possible to “triangulate” or combine several different views on the same events.

Interviews were conducted after two weeks of hospice care. This timing allowed us to gather family members’ perspectives on previous events while also exploring their expectations for the future. Because the length of time from the onset of symptoms to the terminal stage of an illness can range from weeks to years, we used a synchronie or “freeze-t\ime” perspective to collapse the progression of events into a static but rich description of common experiences (e.g., onset of symptoms, diagnosis, and treatment decisions), regardless of the amount of time that had elapsed (34).

Recruitment

A letter describing the study was included in the home care admissions packets of the participating hospice. Telephone contact was made with all who met the study inclusion criteria, which were: patient age 50 years or older, palliative performance scale (PPS) score of 40-50, and at home after two weeks of hospice care. A PPS score of 40-50 indicates that a person requires considerable assistance with self-care but is able to converse (35). Patients with lower PPS scores; those who fatigued easily, who were confused, or who lacked capacity to consent were excluded. Appointments were made with interested participants at a time and location of their choice. Theoretical sampling, which involves selecting respondents who can describe their reactions to the conditions being explored, was used and recruitment continued until new patterns of response had ceased (42,43).

Interviews

When possible, joint interviews were conducted with family members and patients. The primary author and two research assistants served as interviewers. The interviewers had worked as medical social workers and were experienced in making home visits, but were not hospice employees. The interviewers attended a four-hour study orientation and observed the principal investigator before interviewing alone.

Semistructured interviews were conducted using an interview guide that began with medical history questions (e.g., onset of symptoms, diagnosis, treatment, and hospitalizations) as well as questions about health care providers. Participants were encouraged to proceed at a pace and direction of their choice and comfort. Open-ended questions were followed by probes that were aimed at generating in- depth descriptions. Interviews lasted between one hour and two and a half hours.

Analysis

Interviews were tape recorded, transcribed, and entered into Non- numerical Unstructured Data Indexing, Structuring and Theorizing (NUD*IST) software. The iterative process of exploring narratives involved several steps of coding and data reduction. First, multiple readings and open coding were used to describe what happened and how families reacted. During this stage, two coders independently created written summaries of family dynamics and developed codes to reflect the central characteristics. One coder had not conducted interviews. This outside perspective was important for decreasing bias and enhancing the rigour of the analysis (34).

Next, intensive coding was used to develop and refine the categories of family response (36,37). Also described as “axial coding”, this technique is used to develop an overall explanatory scheme (38). Six modes of patient-family dynamics were distilled. The concept of a mode was chosen to characterize different patterns or styles of behavioural responses adopted by those interviewed for managing the distressing events of a progressive illness (39). A text file was created which contained illustrative and descriptive quotes. Cross-case analysis was performed to isolate patterns by creating a matrix to summarize and characterize emergent responses (36). Using this grid as an organizing tool, the salient features of different responses were distilled into six modes. The final step of the analysis involved reexamining the narratives to explore whether fami lies moved between modes and, if they did, how.

RESULTS

Sample

Interviews were conducted with members of 70 families. Table 1 provides a visual representation of the different types of interviews (individual or joint). The total number of participants was 108 (75 family members and 33 patients). The sample was primarily Caucasian (92%), with 7% African American and 1% Hispanic participants. Participants’ religious affiliations were Catholic (46%), Protestant (31%), Jewish (3%), other (7%), not active (6%), and none (7%).

Patients ranged in age from 54 years to 88 years (mean: 74 years). Thirty-three (46%) were men and 38 (54%) were women. Patients’ diagnoses were primarily cancer (81%); noncancer diagnoses included congestive heart failure (7%) and lung conditions (4%), dementia (3%), and debility unspecified or multiple comorbid conditions (4%). Family members’ ages ranged from 21 years to 87 years (mean: 56 years) (Table 2).

Table 1 / TYPES OF INTERVIEWS

Six Family Modes

The six family modes are:

1. reactive: illness generates distress;

2. advocacy: vulnerability ignites assertive action;

3. fused: illness and decline are shared;

4. dissonant: diametrically opposed viewpoints cause struggle;

5. resigned: decline and death are anticipated; and

6. closed: outward responses are impassive.

In this paper, the modes are described, illustrated with participants’ words, and summarized (Table 3).

Reactive Mode: Illness Generates Distress. This mode combines some of the following features: heightened sensitivity, intense emotional expression (e.g., frequent tears, angry outbursts), feeling tense and uneasy about the rapid or dramatic changes of advanced disease, and family conflict. Behavioural responses included sleep disturbances, appetite changes, and a pervasive sense of apprehension, which limited participants’ abilities to focus on issues other than the illness. Participants also expressed feelings of intolerance for other peoples’ idiosyncratic behaviours. For example, one person’s increased need for repeated explanation of medical information generated frustration in other family members; this would have been overlooked or tolerated in other circumstances. In other situations, the ill person became unable to manage personal care tasks; participants in the reactive mode expressed feelings of awkwardness about providing physically intimate care (e.g., assistance with bathing).

Several features of the reactive mode can be illustrated through the “M” family’s experiences. While Mrs. M was hospitalized with advanced pancreatic cancer, her family became concerned about her intermittent confusion. Mrs. M misunderstood their attempts to help, angrily asking, “How do you think you know what I need and want?” A flurry of hurt feelings and conflict followed. After her discharge, she wanted only her daughters to provide hands-on personal care, despite her sons’ attentive presence, and this led to misunderstanding between her children. The following quote illustrates this tension:

My sister and I were plugging in her feeding tube, dealing with her medicine and diet. She didn’t want anyone else to touch her. At one point she said, “I don’t want him (my brother) to do it.” He was standing right there. It was awful.

Features of the reactive mode were described in combinations that were characteristic of each family’s unique context. Influenced by age; stage of life; and racial, cultural, and religious values, their emotional and behavioural expressions reflected collective family distress. Most families experienced the reactive mode at turning points such as: diagnosis, terminal prognosis, medical crises, and when major functional decline occurred. Participants’ descriptions indicated that the intensity of their reactions subsided after the crisis had passed, but families experienced subsequent reactivity with subsequent changes. The reactive mode became the outward expression of internal reactions to major changes.

Table 2 / SAMPLE DEMOGRAPHICS

Table 3 / FAMILY MODES OF RESPONSE

Advocacy Mode: Vulnerability Ignites Assertiveness. This mode is a combination of the following features in one or more family advocates: assertive behaviour which is aimed at improving the patient’s care from providers; verbalizing the person’s needs for pain control or symptom management; and insisting on the patient’s rights for information, dignity, or respect. Advocate(s) spoke with health care or social service professionals when the ill person was unable to voice his or her needs. Some advocates voiced the patients’ unmet needs and challenged providers or administrators with inadequacies in the healthcare system. Others negotiated with physicians and their office personnel, kept records, wrote letters, and sometimes became demanding. Some family members established their presence as an advocate more quietly, by attending appointments, staying overnight in the hospital, and talking with providers regularly. Advocacy was also seen in a subset of older women who discovered (with surprise) their previously dormant ability to speak out as they observed poor treatment of their spouse. One of Mrs. D’s four daughters described her advocacy efforts during a hospitalization:

The doctor called to say she could go home. I told him she wasn’t ready and he said, “If you can’t take care of her, put her in a nursing home.” It was a snowy winter night and I refused. I called the social worker in the morning and said, “If she is really doing great, we’ll take her home. But how do you expect her to be alone if she can’t walk from the bed to the couch?” She said, “Dr. D. said she looked good.” I said, “But she can’t walk.” So she said, “Give me your work number, we’ll watch her walk.” She called me back, apologized, and said, “You’re right.”

Advocacy primarily occurred during the stages of illness progression, if caregivers’ encounters with formal healthcare or social service providers were cold or inappropriate, or if the providers treated the ill person as only a diagnosis.

Fused Mode: Illness Is Shared. This mode combines the following features: the illness became the central focus; all activities centred on the patient’s daily routines, symptoms, food intake, doctors’ appointments, and physical needs for care; all other family issues were minimized. The illness was described as a shared experience by, for example, saying “we” have cancer, leukemia or \heart disease. In addition, participants in the fused mode made statements such as, “We are having surgery”, , “…chemo”, “…radiation”, or “…trouble eating.” This phenomenon was explained by one family member who said, “I say ‘we’ because my father and I share everything. He’s first, I’m second. This is my return for everything he did for me.” Fused emotional responses to the illness were also described in a shared context, which can be illustrated by a daughter who said, “We cry together.” The fused mode is illustrated by Mrs. F’s description of her husband’s chemotherapy:

I think we started the chemo in July. We were having that and then we’d go out for breakfast. He was feeling great. Then they put us on [a new chemotherapy drug] which was brutal. We’d go three Wednesdays a month and then a week off. After the first month, it was diarrhea all the time, terrible side effects, and heartburn. Finally he told the oncologist, “The fact is, we just can’t do this anymore.”

Families in the fused mode reacted to threatening changes by facing each one as a team. In some situations, formal caregivers spoke for the terminally ill person and had vicarious experiences of their loved one’s discomfort, pain, and emotions.

Dissonant Mode: Opposed Viewpoints Cause Struggle. This mode was characterized by the simultaneous presence of diametrically opposed views on the illness and different styles of communication. Two possible scenarios were present: (1) the family was open but the patient closed, or (2) the patient was open but at least one family member was closed to dealing with the illness and decline. Strained communication was a prevalent theme expressed by families in this mode.

When the family was open to and aware of the needs generated by the person’s illness and decline but the patient was not, dissonance erupted around safety issues (e.g., falls, smoking), the introduction of hospice care or of devices such as a hospital bed or bedside commode, or final planning issues. In one caregiver’s words:

He still wants to assert his independence. Today he insisted on taking a shower-in the shower. I said, “I’ll go get the shower chair.” He said, “You bring that thing up here and I’m going to throw it out the window.” I said, “I’m going to stand next to the shower and make sure you don’t fall. I just want you to be safe, that’s all.”

Decline-imposed changes became difficult and humiliating waypoints on the illness trajectory for the ill person who fought to retain independence.

When the patient was open but one or more family members did not want to discuss the situation, a different type of dissonance resulted. A segment of a joint interview is illustrative:

Wife: There are some times when he wants to talk but I don’t. What do you talk about when you’re in your house together all the time? You sit on the porch. We run to the store, and to get the medicine, and that.

Husband: We don’t talk about it much at all. I worry so much about what she’ll need to do after I’m gone but I can’t say to her, “I know that this and that needs to be done”, because she don’t like to discuss it. I don’t want to leave her not knowing what to do.

Families in the dissonant mode may exhibit conflicted communication. This behaviour illustrates how people develop awareness of another’s terminal condition in their own distinct time.

Resigned Mode: Decline and Death are Anticipated. This mode was characterized by acceptance and resolution. Patients whose families were in the resigned mode were most often of advanced age or had been dealing with a life-limiting illness for years. Families understood that there was no cure for the disease, and viewed death as a natural and sequential event. Emotional reactivity was present and integral but accompanied by the anticipation of death. The resigned mode can be illustrated by Mr. N whose mother had a stroke and leukemia. He said:

My mother is 81 years old, she’s lived a good long life and she will have a pain-free death. What more could you ask for? That seemed to brighten her spirits because Mom has never been afraid to die-she knows where she is going.

The approaching death was the naturally occurring resolution of a long-term illness and decline.

Closed Mode: Impassive Outward Response. This mode combined the following features: minimal expression of emotion, bad news handled matter-of-factly, limited questions about or discussion of crisis events. The lack of outward signs or indications of emotional expression about the illness make this mode conceptually different from the resigned mode. In this mode, family members avoid reactivity. One family member described this process by saying, “You don’t stop to think about it; you just keep going and put one foot in front of the other.” Features of the closed mode are illustrated by this quote from Mr. V’s daughter:

We haven’t told my father about his lung tumour. We didn’t think it was necessary. He knows he’s in bad health anyway. His emphysema…is gonna kill him, [the tumour] is just something more. We’ve never actually kept it from him. We just don’t discuss it because he sits around the house worrying about his ailments all day long anyway so it’s just one more thing.

Table 4 / CHANGES THAT TRIGGER MOVEMENT TO A DIFFERENT MODE

The closed mode may result from emotional numbness or submerged feelings. Interviews with families in the closed mode were shorter, answers from both the ill person and caregiver were brief and accompanied by limited description.

Trigger Events

The turning points of a progressive illness changed family dynamics and triggered new modes of response. Three types of events moved families from one mode to another. Functional changes were cognitive, physical, or social (role reversal). Physical transitions included the decreased ability to walk independently; get in and out of bed; or shower, dress, and eat independently. Cognitive changes sometimes came on suddenly, rendering the person unable to communicate clearly. Social changes resulted from both physical and cognitive factors, increased the need for assistance in instrumental activities of daily living, and changed roles. Crisis events were dramatic medical or symptomatic changes such as a stroke, uncontrolled bleeding, or a hospital admission. Provider communication which resulted in new diagnostic or prognostic information signalled increased awareness of the terminal nature of the patient’s condition and, thus, permanent change. Each type of trigger event was accompanied by uncertainty, as well as distinct types of losses. Table 4 summarizes the trigger events.

DISCUSSION

The death of a loved one is a universal experience, however, there is great variability in the ways in which families react to the progression of a life-limiting illness. An analysis of in-depth interviews with 108 people (hospice patients and their family members) yielded detailed information about how families function on the downhill trajectory of a life-limiting illness; a typology of six modes was developed to describe families’ responses. Data from these analyses suggest that family responses are not fixed, and movement from one mode to another is triggered by the ill person’s functional changes, crisis events, and new medical or prognostic information.

Families’ experiences parallel those of the ill person’s on the trajectory of a life-limiting illness. Their associated responses emerge from a unique and distinct combination of universal illness events and unpredicted occurrences that make each family’s trajectory unique. The results of this study build on previous research (14,24) which has documented

* universal events: onset of symptoms, diagnosis, prognosis, decisions about the beginning and endpoints of treatment, advancing illness, physical decline; and

* unforeseen events including: hospitalization, the onset of sudden and dramatic changes (e.g., bleeding, seizures), uncontrolled pain or other symptoms, and the variable quality of communication with providers.

Both universal and unpredicted events cause irrevocable change and new dynamics as the family seeks to retain stability. Life- limiting illness becomes both an internal and external threat to homeostasis, causing loss of control, limited visibility of the future, a need for guidance, altered communication, and growing awareness that the family will never be the same. Individuals vacillate between hope and despair, uncertainty, and fear of the unknown. These internal processes appear outwardly as the family’s mode of response.

Family responses to a loved one’s life-limiting illness are influenced by the age and life stage of the ill person; family members’ racial, ethnic, and cultural differences; and prior experiences with illness and loss. Thus, the same event may generate different modes of response for different families. Unique clusters of behaviours characterize each mode. Families in the reactive mode experience many coexisting emotions, sometimes rapidly changing, for example, sadness, anger, pride, and joy. In the advocacy mode, families may appear angry or hostile because they present an outwardly forceful or assertive stance. It is important to note that this may reflect fear and distress from watching a loved one’s decline or dissatisfaction with the sometimes-impersonal medical system. Families in the dissonant mode experience underlying frustration, which results from conflicting individual responses to the illness, thus creating conflict and communication barriers. This can cause an emotional divide between loved ones. In the fused mode, families join together as a way of managing their shared distress; this can be viewed as a protective response. Families in the resigned mode experience profound sadness, which accompanies the realization that the illness cannot be cured and will end in death. Although families in the closed mode may appear outwardly unemotional, this may buffer inte\nse emotions that are difficult to express.

The results of this study both build on and add to knowledge about family dynamics in a life-limiting illness. Prior studies have illustrated that an advancing illness is accompanied by anticipatory grief, uncertainty, and anguish (7,8,18,39). Building on knowledge about this type of family crisis, the present study reveals variable responses to these intense experiences. Understanding differential responses is key to providing individual and family-centred care (10).

Extending the work on the importance of patient-centred, family- focused care (11), this study illustrates some of the family dynamics which practitioners encounter. The dynamics illustrated within each of the six modes advance understanding of how families grapple with the major changes and approaching losses that accompany a progressive illness. For example, the dissonant mode illustrates how individuals in a family arrive at the same conclusion (the terminal nature of the illness), but do so at different times and in their own ways. The interactions between family members and healthcare professionals are central mechanisms by which families make sense of the complex changes that are occurring. Trigger events create opportunities for meaningful interaction with providers, which can further a sharpened awareness of dying; thus, they present important opportunities for helping families navigate unforeseen change.

The results of this study have implications for health care providers who encounter families at the significant turning points of an advancing illness. Health care providers offer meaningful and focused situation-specific assistance to families. The significance of family-provider interactions throughout the course of an illness cannot be underestimated, and these results further suggest the importance of helping families anticipate the potential endpoints of treatment. The delivery of bad news can be met with intense emotions, angry or forceful behaviour (advocacy mode), or an extremely tearful and emotional response (reactive mode), and this can be alarming to a health care provider. Provider communication can be hindered by fears of causing psychological harm, and by limited awareness of now to assess families’ needs, concerns, and feelings (44). Straightforward but sensitive and unhurried communication has been found to help families ask questions and talk about possible outcomes, including the eventual assistance of palliative care (16,43,44). Anticipatory guidance can be provided by professionals from all disciplines and may help a family normalize the intensity of their responses.

This study had several limitations. First, it is important to note that the hospice patients in many families were unable to participate. The ill person’s view is a critical component of the family perspective and, although described by relatives, its absence renders the family’s story incomplete. This limitation underscores both the importance of gaining the patient’s account and the inherent difficulties of doing so within the illness context. Further, the family perspective was drawn on the accounts of as many people as were willing to participate, however, in all families, some perspectives were missing. This limitation further underscores the importance of understanding the response of the whole family system and the difficulty in doing so. Finally, although interviews were long and detailed, only one was conducted with each participant. A longitudinal view, provided by adding even a second interview, would have yielded a deeper perspective. Conducting longitudinal research at the end of life is important for understanding the final changes that families undergo as a loved one dies, while it runs the risk of being intrusive during these important final moments. These limitations underscore the importance of seeking family perspectives in research about life-limiting illness and of doing so in a sensitive but noninvasive manner.

The importance of future research into family responses to an advancing illness is illustrated by the results of this study. The six modes presented here are a first step toward understanding the cumulative effect of illness events, and these results indicate the importance of future research to refine these modes and examine them in different situations. Additional investigation of how families are helped at the turning points of life-limiting illness and the long-term outcomes of this assistance can inform both clinical education and practice. Providers who understand the clinical conditions they treat, as well as the varied responses they observe, can improve the delivery of end-of-life care.

Date received, April 5, 2004; date accepted, September 30, 2004.

ACKNOWLEDGEMENTS

The authors gratefully acknowledge the ongoing support of William Finn, CEO of the Center for Hospice and Palliative Care, Buffalo, New York, and the editorial assistance of Drs. Nancy Hooyman, Deborah Padgett, and Daniel Gardner, who critiqued earlier versions. This assistance was facilitated by the professional network that has been established through the John A. Hartford Foundation Faculty Scholars Program.

Funding for this project was provided by the Margaret L. Wendt Foundation of Buffalo, New York, USA.

REFERENCES

1. Kristjanson LJ, Leis A, Koop, PM Mueller B. Family members’ care expectations, care perceptions, and satisfaction with advanced cancer care: results of a multisite pilot study. J Palliat Care 1997; 13(4): 5-13.

2. Weitnzer MA, Knutzen R. The impact of pituitary disease on the family caregiver and overall family functioning. Psychother Psychosom 1998; 67: 181-188.

3. Brown MA, Powell-Cope GM. AIDS family caregiving: transitions through uncertainty. Nurs Res 1991; 40: 338-345.

4. Craddock RB. Commentary on AIDS family caregiving: transitions through uncertainty. ONS Nurs Scan Oncol 1992; 1: 7.

5. Zhang AY, Siminoff LA. Silence and cancer: Why do fami lies and patients fail to communicate? Health Commun 2003; 15: 415-429.

6. Glaser B, Strauss A. Awareness of Dying. New York: Aldine, 1965.

7. Adelbratt S, Strang P. Death anxiety in brain tumour patients and their spouses. Palliat Med 2000; 14: 499-507.

8. Cherny NI. The treatment of suffering in patients with advanced cancer. In: Chochinov HM, Breitbart W (eds). Handbook of Psychiatry in Palliative Medicine. Oxford: Oxford University Press, 2000; pp. 375-396.

9. The Task Force on Palliative Care: Precepts of Palliative Care. Available at http://www.lastacts.org, 1997.

10. Emanuel EJ, Fairclough DL, Slutsman J, Emanuel LL. Understanding economic and other burdens of terminal illness: the experience of patients and their caregivers. Ann Intern Med 2000; 132: 451-459.

11. Teno JM, Casey VA, Welch LC, Edgman-Levitan S. Patient- focused, family-centered end-of-life medical care: views of the guidelines and bereaved family members. J Pain Symptom Manage 2001; 22: 738-751.

12. Weitzner MA, McMillan SC, Jacobsen PB. Family caregiver quality of life: differences between curative and palliative cancer treatment settings. J Pain Symptom Manage 1999; 17: 418-428.

13. Kissane DW, Bloch S, Burns WI, Patrick JD, Wallace CS, McKenzie DP. Perceptions of family functioning and cancer. Psychooncology 1994; 3: 259-269.

14. Enyert G, Burman, ME. A qualitative study of self- transcendence in caregivers of terminally ill patients. Am J Hosp Palliat Care 1999; 16: 455-462.

15. Powazki RD, Walsh D. Family distress in palliative medicine: a pilot study of the Family APGAR Scale. Am J Hosp Palliat Care 2002; 19: 392-396.

16. Byock I. Completing the continuum of cancer care: integrating life-prolongation and palliation. Cancer 2000; 50: 123-132.

17. Lederberg MS. The family of the cancer patient. In: Holland JC (ed). Psycho-oncology. New York: Oxford University Press 1999; pp. 981-993.

18. Loscalzo M, BrintzenhofeSzoc K. Brief crisis counseling. In: Holland JC (ed). Psycho-oncology. New York: Oxford University Press 1999; pp. 662-675.

19. McCorkle R, Pasacreta JV. Enhancing caregiver outcomes in palliative care. Cancer Control 2001; 8: 36-45.

20. Meares CJ. Primary caregiver perceptions of intake cessation in patients who are terminally ill. Oncol Nurs Forum 1997; 24: 1751- 1757

21. Doyle-Brown M. The transitional phase: the closing journey for patients and family caregivers. Am J Hosp Palliat Care, 2000; 17: 354-357.

22. Olson K, Morse HM, Smith JE, Mayan MJ, Hammond D. Linking trajectories of illness and dying. Omega 2000; 42: 293-308.

23. Pincombe J, Brown M, McCutcheon H. No time for dying: a study of the care of dying patient in two acute care Australian hospitals. J Palliat Care 2003; 19: 77-86.

24. Mehta A, Ezer H. My love is hurting: the meaning spouses attribute to their loved ones’ pain during palliative care. J Pallit Care 2003; 19: 87-94.

25. Andershed B, Ternestedt B. The illness trajectory-for patients with cancer who died in two different cultures of care. Omega 1998; 37: 251-272.

26. Sherman DW. The reciprocal suffering of caregivers. In: Doka KJ, Davidson JK (eds). Caregiving and Loss: family needs, professional responses. Washington, DC: Hospice Foundation of America, 2001; pp. 247-264.

27. Brown MA, Stetz K. The labor of caregiving: a theoretical model of caregiving during potentially fatal illness. Qual Health Res 1999; 9: 182-197.

28. Cameron Jl, Franche RL, Cheung AM, Stewart DE. Lifestyle interference and emotional distress in family caregivers of advanced cancer patients. Cancer 2002; 94: 521-527.

29. Stajduhar KI, Davies B. Palliative care at home: reflections on HIV/AIDS family caregiving experiences. J Palliat Care 1998; 14: 14-22.

30. Abrahm JL. Update in palliative medicine and end-of-life care. Ann Rev Med 2003; 54: 53-72.

31. Walsh F, McGoldrick M. Loss and the family. In: Walsh F, McGoldrick M (eds). Living Beyond Loss. New York: Norton, 1991; pp. 1-29.

32. Wein, S. The family in terminal illness. In: B\aider L, Cooper CL, Kaplan De-Nour A (eds). Cancer and the Family. Hoboken, New Jersey: John Wiley & Sons, 2000; pp. 427-441.

33. Addington-Hall J, McPherson C. After-death interviews with surrogates/bereaved family members: some issues of validity. J Pain Symptom Manage 2001 ; 22: 784-790.

34. Padgett DK. Qualitative Methods in Social Work Research. Thousand Oaks, California: Sage, 1998.

35. Anderson F, Downing GM, Hill J. Palliative performance scale (PPS): a new tool. J Palliat Care 1996; 21: 5-11.

36. Miles MB, Huberman AM. Qualitative Data Analysis. Thousand Oaks, California: Sage, 1994.

37. Cresswell JW. Research Design Qualitative & Quantitative Approaches. Thousand Oaks, California: Sage, 1994.

38. Strauss A, Corbin J. Basics of Qualitative Research Techniques and Procedures for Developing Grounded Theory. 2nd edition. Thousand Oaks, California: Sage, 1998.

39. Yedidia MJ, MacGregor B. Confronting the prospect of dying: reports of terminally ill patients. J Pain Symptom Manage 2001; 22: 807-819.

40. Yates P, Stetz KM. Families’ awareness of and response to dying. Oncol Nurs Forum 1999; 26: 113-120.

41. Lynn J. Measuring quality of care at the end of life: a statement of principles. J Am Geriatr Soc 1997; 45: 526-527.

42. Maguire P. Communication with terminally ill patients and their relatives. In: Chochinov HM, Breitbart W (eds). Handbook of Psychiatry in Palliative Medicine. New York: Oxford University Press, 2000; pp. 291-302.

43. Grbich CF, Maddocks I, Parker D. Family caregivers, their needs, and home-based palliative cancer services. J Fam Stud 2001; 7: 171-188.

44. Davies B, Reimer JC. Family functioning and its implications for palliative care. J Palliat Care 1994; 10: 29-36.

DEBORAH P. WALDROP, University at Buffalo School of Social Work, Buffalo, ROBERT A. MILCH and JUDITH A. SKRETNY, The Center for Hospice and Palliative Care, Cheektowaga, New York, USA

Copyright Center for Bioethics, Clinical Research Institute of Montreal Summer 2005