Sources of Stress for Hospice Caregiving Families
Margaret M. Hull
SUMMARY. The purpose of this qualitative study was to generate a detailed description of the concerns and stresses families encountered as they cared for their dying relative in an oncology hospice home care program. Semi-structured interviews and participant observation were used to follow 14 family members from 10 different families throughout their caregiving experiences. The constant comparative method of data analysis was used to discover emergent themes. Families identified three general sources of stress: patient symptoms, interactions with others, and concerns for self. Findings were examined from within the larger body of caregiving literature. Changes in their relatives’ mental status seemed most stressful over time, and is an area that may benefit from increased support from health professionals.
Hospice home care of a dying relative is a rapidly growing alternative to traditional hospital based terminal care for patients with cancer. The number of hospices in the United States increased from one in 1971 to currently over 1700 fully operational programs (National Hospice Organization [NHO], 1988). With the predominant focus of home care, families play a vital role in providing the day to day care for their dying relatives. Within the past decade, family caregiving has been the focus of major research attention. Areas of investigation include (a) characteristics of caregivers and care recipients; (b) tasks of caregiving; (c) impact of caregiving on the caregiver, especially the frail elderly; and (d) impact of caregiving for a relative with senile dementia or Alzheimer’s disease. Except for studies that focused on caring for the cognitively impaired, caregiving has not been examined within specific medical diagnoses. It seems that some needs, concerns, difficulties and stresses may be closely related to the course and treatment of a given illness and that caring for a dying relative with a cancer versus a non-cancer illness might be a different experience. In addition, hospice care is offered for the last 6 months of life. Examining caregiving that involved a relatively short term commitment rather than the longer involvements typically noted in the caregiving literature may add information to this body of knowledge. However, far less is known about the caregiving experiences of hospice home care families whose relative is dying of cancer. Therefore, examination of caregiving within hospice as a home care alternative seems warranted.
Previous nursing research with families caring for a terminally ill relative at home has focused on two variables, family needs and the supportiveness of nursing behaviors. Of most importance to families was knowledge about their relatives’ conditions, prognosis, and signs of imminent death (Garland, Bass, & Otto, 1984; Googe & Varricchio, 1981; Skorupka & Bohnet, 1982). Assurance about their relatives’ comfort and adequacy of pain control also were of primary concern (McGinnis, 1986; Skorupka & Bohnet). Learning needs regarding these measures as well as ambulation techniques were identified by home care families (Grobe, Ilstrup, & Ahmann 1981; Hinds, 1985).
Within these studies families were asked to respond to multiple item inventories developed by the researchers. Content validity was well established through the use of extensive literature reviews and consultations with experts in the areas of oncology, death and dying, loss, and hospice care. However, families’ experiences were rarely used to generate descriptive items included in the research tools. Rather, families were asked to respond to statements others felt to be appropriate for them during their relatives’ terminal illnesses. Though these studies were helpful in initial identification of caregivers’ areas of concern, the cross-sectional approach limited understanding of this clinical problem to one point in time during the terminal period. Since the duration of the terminal experience encompasses more than a single instance, it is expected that the perspectives of family caregivers will change over time and be affected by personal, environmental, and social influences related to their own situation and also their relatives’ changing conditions. Use of a cross-sectional design has not addressed the dynamic nature of this experience over time, within the social context in which it occurs, and limits the ability to understand the full nature of what families deal with. Allowing family members to generate this information from their own life situations as they go about caring for their dying relatives at home will enhance our understanding of what this experience is like for them. Subsequently, health care professionals can begin to develop interventions for further investigation based on families’ accounts of the concerns and stresses they confront as caregivers for their dying relatives. Therefore, the purpose of this study was to generate a detailed description of the concerns and stresses families encountered as they cared for their dying relatives within an oncology hospice home care program.
METHOD
A longitudinal qualitative methodology combining semi-structured interviews and participant observation was used to examine the subjective reality of this experience from the participants’ points of view. Multiple home visits were conducted with each caregiving family in order to observe caring within the context in which it occurred. Frequency of visits was determined primarily by changes in patients’ conditions. If there was stability in the caregiving experience, visits were made every 3 to 4 weeks. If changes in patients’ conditions occurred, or changes in caregivers’ circumstances or experiences were indicated, more frequent visits were made. Phone contact with the hospice nurses as well as with families was used to gather information that could be used to assess changes that warranted a visit.
Setting and Sample
This study was conducted in an oncology hospice program in a large northeastern city. Home care was the major focus of this program, though five inpatient beds were available at a free-standing unit (FSU). This facility was used primarily to offer short term respite for caregivers or for acute symptom control. Hospice census was maintained at 30–35 patients. To be eligible for the study, family members had to have a relative enrolled in the hospice program, be 18 years of age or older, and able to understand and speak English. Participants could be the primary caregiver in the home, any other adult living in the home with or without care responsibilities for their dying relative, or any adult living outside the home who had consistent patient care responsibilities. The variety of family members was an attempt to examine the extent to which the caregiving experience impacted on all family members, rather than a single primary caregiver.
Staff nurses introduced the study to eligible families on a routine home care visit and left a copy of the consent. This consent contained detailed information on the purpose of the study, what participation involved, and the risks and benefits of participation. The investigator made phone contact with families within two days to discuss any questions and elicit participation.
Fourteen individual family members from 10 different families were enrolled. Seven families contributed a single family member, the primary caregiver (PCG). Two families contributed two family members, the primary caregiver and her spouse. One family contributed three family members, a father and son who alternated the role of PCG, and a daughter who lived out of state and served as PCG during periodic visits over the course of her mother’s illness. Relationship of caregivers and patients varied. Six children were caring for a parent, 4 wives and 1 husband were caring for their spouse, 2 sons-in-law were helping to care for their wife’s parent, and 1 niece was caring for her aunt. Ages of caregivers ranged from 26 to 78. Ages of patients ranged from 41 to 90. Table 1 displays ages and relationships of caregivers and their dying relatives.
TABLE 1.
Age and Relationship of Caregivers and Patients
| AGE and RELATIONSHIP | |
CAREGIVER | | PATIENT |
40 year old | wife | caring for | 41 year old | husband |
49 | wife | | 54 | husband |
75 | wife | | 85 | husband |
78 | wife | | 85 | husband |
64 | daughter | | 83 | mother |
40 | niece | | 69 | aunt |
26 | son | | 65 | father |
40 41 | daughter & son-in-law | | 75 75 | mother/ mother-in-law |
49 50 | daughter & son-in-law | | 90 90 | father/ father-in-law |
75 35 37 | husband, son & daughter | | 70 70 70 | wife/ mother/ mother |
Procedures for Data Collection
At the first interview written informed consent was obtained from each participant. Permission to tape record the inteiviews was given by all subjects. All data were kept confidential. Interviews were conducted in the home. Privacy was easily maintained because patients were out of the home for short periods, napping, or in their rooms. Caregivers were assured that interruptions to maintain patient care responsibilities were expected and would not be a problem for the investigator. This was often a time when participant observation was possible.
When more than one family member participated in the study, the first interview was conducted with both subjects present. Families seemed more comfortable together for this initial visit. Subsequent interviews were conducted individually to encourage more open communication.
Specific questions were addressed on the first interview. These included: What concerns do you have that are specifically related to caring for your dying relative? In what areas of your life do you feel you need more help/understanding/information? What are the most difficult parts about caring for your relative at home? What symptoms are most difficult for you to deal with? Once structured content on the interview guide had been addressed, following interviews were more open-ended. Initial interviews were 2–3 hours after which a flexible schedule of informal visits and interviews was established with each family. Subsequent interviews averaged 60–90 minutes. Field observations were taped immediately after each visit to assure accuracy and thoroughness of the investigator’s impressions. Typed transcriptions of interviews and field observations were made and clarified by listening to the tapes while reading the transcriptions.
A final interview was conducted 3 weeks after the patient died and served several purposes: it allowed the researcher to extend her sympathies for the family’s loss in person, thank them for sharing their experience, and establish closure. Final interviews were conducted with 13 family members in 9 of the 10 study families: 12 family members whose relative died, and 1 son who withdrew his father from hospice to pursue a day care program of occupational and physical therapy. One patient was alive at the end of data collection and continued in the hospice program under his wife’s care.
In summary, 55 visits were made over 16 months of data collection. Total time families were enrolled in hospice ranged from 2.5 to 31 months. Prior to enrollment in the study, families had been involved in the hospice program anywhere from 1 to 26 months. Participation in the study lasted from 1 to 12 months. Visits per family ranged from 2 to 10.
Data Analysis
The constant comparative method of content analysis of transcribed interviews and field notes involved coding for themes emerging over the course of data collection (Glaser, 1978; Glaser & Strauss, 1967). Phrases, sentences, or anecdotes that represented individual thoughts, feelings, or concepts were separated into relevant themes. Coded data were clustered into categories that described and explained phenomena. Categories were compared with each other to discover conceptual and theoretical linkages between them. Related categories were merged and reduced and new categories were added. Saturation of data occurred when new categories ceased to emerge and no different responses or observations were obtained. Throughout this process codes, i.e., conceptual labels, were assigned to data at all levels of abstraction. Recurrent themes became evident as single words or short phrases were coded and analyzed for similar content. Often, subjects’ words or ideas were used to identify this latter group of categories.
Analytic notes had been made throughout data collection and reflected thoughts and questions suggested by the coding process, thus allowing openness to additional ideas. These notes served several functions. First, they facilitated recording and organizing results of the analysis. Second, analytic notes indicated what ground the investigator had covered and what directions were necessary to explore as the research progressed. Finally, analytic notes were the basis for writing the findings (Corbin, 1986).
Validation and Verification of Data
Since the purposes, goals and world views of qualitative research differ from those of the more traditional quantitative research, several authors have suggested more appropriate standards by which to evaluate the scientific merit or rigor of qualitative studies. Sandelowski (1986) suggested several strategies for achieving rigor in qualitative research based on the work of Guba and Lincoln (1981). Four factors compose this framework: credibility, applicability, consistency, and confirmability.
Credibility, or truth value is established when those individuals who were the focus of the investigation immediately recognize themselves and their experience in the descriptions and interpretations of the phenomenon. In addition, credibility occurs when readers other than the researcher or the subjects recognize the described experience. In this study, credibility was addressed in two ways. First, subjects were asked to comment on the “truth” of the findings as described and to confirm interpretations the researcher made from their experiences. Second, several individuals not involved in this study were asked to review the findings and comment on their ability to recognize the described experience.
Applicability or fittingness, refers to the ability to “fit” findings into the data from which they origin...