Informal Caregiving of Hospice Patients

Abstract

Background: Informal caregivers play a critical role in the provision of care to hospice patients. The care they provide often impacts their physical and psychological well-being.

Objective: This study synthesized 58 articles pertaining to informal hospice caregiving, focusing on caregivers' satisfaction with hospice services, the physical and psychological well-being of caregivers, the predictors of caregivers' well-being, the direct impact of hospice services on caregivers, and the effectiveness of targeted interventions for hospice caregivers.

Method: A systematic literature review of journal articles published between 1985 and 2012 was conducted.

Results: The studies reviewed found hospice caregivers to experience clinically significant levels of anxiety, depression, and stress; however, results for caregiver burden and quality of life were mixed. Caregivers' perceptions regarding the meaningfulness of care as well as their levels of social support were associated with enhanced psychological outcomes.

Conclusions: Beyond satisfaction with hospice services, the direct impact of standard hospice care on caregivers remains uncertain. Caregiver intervention studies have demonstrated promising outcomes signifying a need for additional investigations into hospice-specific interventions that improve caregiver outcomes. Additional research and resources are needed to assist hospice caregivers, with the ultimate goal of minimizing their psychiatric and physical morbidity and enhancing their caregiving and subsequent bereavement processes.

Introduction

Patients with life-limiting illness often wish to remain at home versus being admitted to a hospital, shifting caregiving responsibilities from formal or institutional caregivers to informal caregivers.^1–3 Family and friends as informal or unpaid caregivers provide an average of 66 hours per week of physical, emotional, and practical care during the last year of a loved one's life.^4,5 In the United States over 70% of patients with life-limiting illness receive care from an informal caregiver at the time of their death, and an estimated three million informal caregivers accompany hospice patients through the dying process annually.⁶ The care provided by family and friends represents considerable economic value to the health care system.^2,3,5 However, providing care to a loved one during the dying process involves significant personal, professional, and financial sacrifices,^6,7 and consequently should also be viewed in terms of its personal and societal costs.

Unfortunately, providing care to a family member or loved one at the end of life has also been associated with detrimental effects on caregivers' physical and psychological well-being^8–14 and increased risk of mortality.^12,15–17 Furthermore, the well-being of caregivers and patients are intimately linked. For example, caregivers' well-being often deteriorates as the burden and stress of caregiving increases, putting the patient at increased risk of hospitalization.^11,18,19 Also, psychological distress within a caregiver-patient dyad is correlated, suggesting that the caregiver and patient are interdependent and respond to distress as a unit.^9,20 As caregivers typically survive beyond the patient, reducing psychological and physical morbidity in these individuals has a long-term impact both on their well-being and their bereavement process after the death of the patient.^11,12,21 For hospice care to be most effective and speak to this public health issue, hospice services must address the needs and stressors associated with informal caregiving.

Objectives

The aim of this study was to review the published research on informal hospice caregivers that evaluated the psychological and physical effect of providing care, the impact of hospice care on caregivers, and the effectiveness of hospice-based interventions for improving caregivers' well-being. Since hospice care is defined differently throughout the world, this review focuses exclusively on caregivers of hospice patients within a U.S. context. The objective of this review is to address the following questions:

• 1) What is the psychological and/or physical well-being of informal caregivers who provide care to a dying family member or friend enrolled in a U.S. hospice program?

• 2) Do hospice services have an effect on caregivers' well-being?

• 3) Do hospice-based caregiver interventions enhance informal caregivers' well-being?

Methods

Design

This study systematically identified and evaluated quantitative and qualitative research on informal hospice caregivers and employed a narrative approach to data synthesis. This approached was selected because of the broad research questions and the diversity of study designs.²² A narrative synthesis was used to assimilate the evidence from the identified studies. This synthesis involved comparing and contrasting the individual studies to identify relevant themes as well as factors that may account for the variability in the results.²³

Data sources

Studies from peer-reviewed medical and psychological journals were extracted by searching three electronic databases: PubMed, PsycINFO, and CINAHL. Searches were conducted by the first author and were limited to studies in English published between January 1985 and December 2012. The following keywords/MeSH search terms were utilized: hospice, caregivers, satisfaction (personal), stress (psychological), caregiver burden, dependency (psychological), well-being, quality of life, health (status), and mental health.

Inclusion and exclusion criteria

Studies were evaluated by two independent reviewers (first and fourth authors) based on an a priori set of criteria; discrepancies in study ratings were settled by a third independent rater (second author). These were the criteria:

• (1) The articles defined an informal caregiver as an individual who provided physical and/or supportive care to a hospice patient, and that care was not provided as part of employment.

• (2) The articles provided a clear description of the methods and results.

• (3) The articles measured outcomes of caregivers' emotional and/or psychological/psychiatric well-being, physical health, or satisfaction.

Because hospice care in the United States involves unique admission criteria and prescribes a core set of services, only caregivers involved in a U.S. hospice program were included. Self-reports, case studies, editorials, as well as studies not published in English were excluded.

Results

Database searches generated 402 articles (see Fig. 1). Abstracts of these articles were reviewed and 292 articles were rejected based on the a priori criteria. The primary reasons for rejection included inadequate caregiver outcomes, a lack of hospice involvement, and participants not receiving services within the United States (i.e., international studies). The complete texts of the remaining 110 articles were evaluated. Of these, 52 were excluded for not clearly distinguishing between palliative care and hospice care, for using single measurement designs, for reporting change based on mean scores without consideration of variability, and/or for not adequately reporting caregiver outcomes. The remaining 58 articles were included in this review.

FIG. 1.

Decision process for inclusion of informal hospice caregiver articles in the review.

Design and quality of reviewed studies

Key methodological features of the studies included in this review are summarized in Table 1. The most common study design (22 studies) was a descriptive quantitative design.^24–29 The next most commonly employed were a prospective quasi-experimental design (15 studies) and a retrospective quasi-experimental design (6 studies).^30–36 Qualitative methodology was employed in 11 studies,^9,11,37–39 and 3 studies employed a randomized controlled trial.^40–42 A few studies were secondary analyses of existing data.^16,28

Table 1.

Key Methodological Features of the Studies Included in this Review of Informal Caregiving for Hospice Patients

Study (reference)	Sample sizea	Percent female; race	Primary disease type; data collection points pre/post patient death	Research design	Response rate (%)	Outcomes
Archer & Boyle, 1999 (24)	44	59% 82% Caucasian, 18% African American	NR Post death	Quantitative descriptive, retrospective	80	Caregiver satisfaction on Primary Caregiver Satisfaction Survey
Bernard & Guarnaccia, 2002 (43)	213	29% 92% Caucasian, 8% Other	Breast cancer Pre & post death	Quasi-experimental, prospective	NR	Grief Experience Inventory and Despair Subscale
Bernard & Guarnaccia, 2003 (30)	213	29% NR	Breast cancer Pre & post death	Quasi-experimental, prospective	NR	Grief Experience Inventory and Despair Subscale
Bialon & Coke, 2012 (10)	9	78% 67% Caucasian, 22% African American	Mixed Post death	Qualitative	NR	Caregiver interviews
Cagle & Kovacs, 2011 (37)	69	72% 84% Caucasian, 4% African American, 11% Other	Cancer Pre & post death	Qualitative, prospective	26	Survey responses
Casarett, Hirschman, Crowley, Galbraith, & Leo, 2003 (52)	112	77% 69% Caucasian, 27% African American, 4% Other	Mixed (predominantly cancer) Post death	Quantitative descriptive, retrospective	54	Satisfaction with services
Chentsova-Dutton, et al., 2000 (32)	181/112	79% 96% Caucasian, 1% African American, 4% Hispanic	NR Pre death	Quasi-experimental, retrospective	NR	Depression (BDI, Hamilton Psychiatric Rating Scale for Depression); Psychological symptomatology (BSI); Health (self-rating estimates, medical visits); Social functioning (Social Support Questionnaire)
Chentsova-Dutton, et al., 2002 (49)	84/48	83% 96% Caucasian, 2% African American, 2% Hispanic	NR Pre & post death	Quasi-experimental, prospective	44	Depression (Hamilton Rating Scale for Depression); Psychological symptomatology (BSI); Grief (Texas Revised Instrument of Grief)
Christakis & Iwashyna, 2003 (44)	61,676	80% 93% Caucasian, others NR	Mixed Post death	Quasi-experimental, retrospective, matched control group	100	Duration of survival of bereaved widows/widowers
Dawson, 1991 (54)	100	66% 95% Caucasian, others NR	NR Post death	Quantitative descriptive, retrospective	58	Three satisfaction measures
Demiris, Oliver, Courtney, & Day, 2007 (74)	12	75% NR	Mixed (predominantly cancer) Pre death	Quasi-experimental, prospective	NR	Anxiety (State Trait Anxiety Inventory)
Demiris et al., 2010 (33)	23	76% 86% Caucasian, 1% African American, 13% Other	NR Pre death	Quasi-experimental, prospective	89	Quality of life (CQLI-R); Anxiety (State Trait Anxiety Inventory)
Demiris, Oliver, Wittenberg-Lyles, & Washington, 2011 (31)	33	81% 98% Caucasian, 2% African American	NR Pre death	Quasi-experimental	79	CQLI-R; State-Trait Anxiety Inventory; Problem-Solving Inventory
Demiris, et al., 2012 (41)	89	79% 87% Caucasian	Pre death	Experimental, prospective	71	CQOLI-R; State-Trait Anxiety Inventory; Problem-Solving Inventory
Empeño, Raming, Irwin, Nelesen, & Lloyd, 2011 (77)	123	82% NR	NR Pre death	Quasi-experimental, prospective	43	Pearlin Role Overload Measure
Evans, Cutson, Steinhauser, & Tulsky, 2006 (55)	18	50% 61% Caucasian, 39% African American	NR Post death	Qualitative, retrospective	22	Caregivers' experiences extracted from semi-structured interviews
Fenix et al., 2006 (68)	175	75% NR	NR Most (63%) post death	Quasi-experimental, prospective	53	Major Depressive Disorder (SCID)
Ganzini, Johnston, & Silveira, 2002 (25)	50	NR NR	ALS Post death	Quantitative descriptive, retrospective	NR	Satisfaction with care (FAMCARE)
Gill, Kaur, Rummans, Novotny, & Sloan, 2003 (59)	57	68% NR	NR Pre death	Quasi-experimental, prospective	NR	CQOL (multidimensional measurement)
Goldstein, et al., 2004 (26)	206	71% 96% Caucasian, 3% African American, 1% Other	Cancer Most post death	Quantitative descriptive, retrospective	53	Caregiver burden (Zarit Burden Inventory)
Goy, Carter, & Ganzini, 2008 (27)	47	77% 92% Caucasian, 8% Other	Parkinson's Post death	Quantitative descriptive, retrospective	39	Caregiver mood (CES-D); Caregiver satisfaction with care
Haley, LaMonde, Han, Burton, & Schonwetter, 2003 (34)	80	79% NR	Lung cancer & dementia Pre death	Quasi-experimental, retrospective	NR	Caregiver depression (CES-D); Life Satisfaction Index
Haley, LaMonde, Han, Narramore, & Schonwetter, 2001 (57)	120	83% 82% Caucasian, 18% Other	Lung cancer & dementia Pre death	Quasi-experimental, retrospective	NR	Caregiver depression (CES-D); Life Satisfaction Index; Caregiver physical health (Health Perception Scale & Medical Outcomes Study Short Form Health Survey)
Herth, 1993 (38)	25	76% NR	Mixed (cancer most common) Pre death	Qualitative and quantitative; descriptive, prospective	NR	Hope (Herth Hope Index and semi-structured interview)
Holtslander & McMillan, 2011 (28)	280	76% 97% Caucasian, 3% African American, 1% Other	Cancer Post death	Quantitative descriptive, retrospective	NR	Depressive symptoms (CES-D); grief (Texas Revised Inventory of Grief); (Inventory of Complicated Grief)
Hull, 1990 (79)	14	79% NR	NR Pre & post death	Qualitative descriptive, prospective	NR	Caregiver stressors (interviews)
Jones, 2010 (64)	388	NR NR	Mixed (predominantly cancer) Post death	Quantitative descriptive, retrospective	NR	Utilization of bereavement services
Kilbourn, et al., 2011 (73)	19	91% NR	NR Pre death	Quantitative descriptive, prospective	NR	CES-D; PSS; MOS SF-36; ENRICHD Social Support Inventory Benefit Finding Scale
Kramer, Kavanaugh, Trentham-Dietz, Walsh, & Yonker, 2010 (71)	152	80% 97% Caucasian, 3% African American	Lung cancer Post death	Quantitative descriptive, retrospective	47	Complicated grief (Inventory of Complicated Grief)
Kreling, et al., 2010 (39)	30	NR 50% Caucasian, 50% Hispanic	NR Post death	Qualitative design, retrospective	NR	Caregivers' experience of hospice (interviews)
Kris, et al., 2006 (50)	175	75% NR	Cancer Pre and post death	Quasi-experimental, prospective	53	Major Depressive Disorder (SCID)
Kutner, et al., 2009 (61)	36	86% 92% Caucasian	NR Post death	Qualitative design, retrospective	NR	Hospice caregivers' needs (interviews)
Kwak, Salmon, Acquaviva, Brandt, & Egan, 2007 (76)	926	81% 88% Caucasian, 7% African American	NR Pre death	Quasi-experimental, prospective	46	Comfort with caregiving tasks; Completion and closure; Caregiving gain (Comfort with Caregiving Scale, Caregiver Closure Scale, CSS)
McMillan, 1996 (46)	118	64% NR	Cancer Pre death	Quasi -experimental, prospective	NR	Caregiver quality of life (CQOLI)
McMillan, et al., 2006 (40)	354	85% NR	Cancer Pre death	Experimental	31 (completed all measures)	Quality of life (CQOL-C); Caregiver burden due to patient symptoms (MSAS); Caregiver burden due to tasks (CDS); Caregiver mastery (6-item scale)
McMillan & Mahon, 1994 (66)	130	85% NR	Cancer Pre death	Quasi- experimental, prospective	NR	Caregiver quality of life (CQLI)
McMillan, Small, & Haley, 2011 (42)	716	74% NR	Cancer Pre death	Experimental, prospective	48%	CES-D; Spiritual Needs Inventory; Mental Status Questionnaire
Meyers & Gray, 2001 (53)	44	75% 91% Caucasian, 1% African American, 8% Other	Mixed (66% cancer) Pre death	Quantitative descriptive, retrospective	94	Satisfaction with hospice care (FAMCARE Scale); Quality of life CQOLC; Burden (Caregiver Strain Index)
Mickley, Pargament, Brant, & Hipp, 1998 (69)	92	78% NR	NR Pre death	Quantitative descriptive, retrospective	37	Depression (CES-D); Anxiety (BAI); Meaning (Purpose in Life Test); Situational Outcomes (General and Religious Outcome Scales)
Moody & McMillan, 2003 (65)	162	78% 85% Caucasian, 11% African American 6% Hispanic, 1% Other	Cancer (mixed cancer types) Pre death	Quantitative descriptive, retrospective	54	Caregiver quality of life
Newton, Bell, Lambert, & Fearing, 2002 (62)	33	73% NR	Mixed (predominantly cancer) Pre death	Quantitative descriptive, retrospective	33	Needs and concerns of caregivers (interview)
Parker, et al., 2010 (75)	24	77% 97% Caucasian	Mixed (predominantly cancer) Pre death	Quasi- experimental, prospective	39	Caregiver quality of life (CQLI-R)
Prigerson et al., 2003 (58)	205	71% NR	Cancer Pre death	Quantitative descriptive, retrospective	61	Caregiver fear and helplessness associated with exposure to death (SCARED scale)
Redinbaugh, Baum, Tarbell, & Arnold, 2003 (80)	31	68% 84% Caucasian, 16% African American	Cancer Pre death	Quantitative Descriptive, retrospective	54	Caregiver strain (BSI, Caregiver Burden Screen)
Salmon, Kwak, Acquaviva, Brandt, & Egan, 2005 (36)	953	77% NR	Mixed (predominantly cancer) Pre post death	Quasi-experimental, retrospective	32	Caregiver gain; Caregiver burden (Revised Caregiving Appraisal Scale)
Speer, Robinson, & Reed, 1995 (67)	160	75% NR	Mixed (predominantly cancer) Post death	Quantitative descriptive, retrospective	30	Caregiver adjustment (Texas Inventory of Grief)
Steele, Mills, Long, & Hagopian, 2002 (56)	443	NR	NR Post death	Quantitative descriptive, retrospective	NR	Satisfaction with hospice services (created scale)
Tang, 2009 (29)	60	65% 97% Caucasian, 3% African American	Mixed (predominantly cancer) Pre death	Quantitative descriptive, retrospective	71	Caregiver quality of life (CQOLC)
Tilden, Tolle, Drach, & Perrin, 2004 (70)	1189	71% NR	Mixed (predominantly cancer) Post death	Quantitative descriptive, retrospective	72	Caregiver strain (4 items)
Townsend, Ishler, Shapiro, Pitorak, & Matthews, 2010 (63)	162	80% 86% Caucasian, 12% African American, 1% Hispanic, 1% Other	Mixed (predominantly cancer) Pre death	Quantitative descriptive, retrospective	NR	Caregiver strain (13-item scale)
Waldrop & Meeker, 2011 (11)	3	81% 94% Caucasian, 3% African American, 3% Hispanic	Mostly Cancer Post death	Qualitative retrospective	32%	Caregiver interview
Weitzner, McMillan, & Jacobsen, 1999 (45)	401	78% 70% Caucasian, 19% African American 9% Hispanic, 2% Other	Cancer Pre death	Quasi-Experimental, retrospective	NR	Caregiver quality of life (CQOLC)
Wilder, Oliver, Demiris, & Washington, 2008 (35)	76	63% 96% Caucasian, 3% African American, 1% Other	Mixed (predominantly cancer) Pre death	Quasi-experimental, prospective	NR	Caregiver quality of life (CQLI-R)
Wittenberg-Lyles, Demiris, Oliver, & Burt, 2011 (60)	30	77% 97% Caucasian, 3% African American	Mixed (not predominantly cancer) Pre death	Qualitative, retrospective	NR	Caregivers' concerns (interviews)
Wittenberg-Lyles, et al., 2012 (9)	81	79% 96% Caucasian	Mixed Pre death	Qualitative retrospective	NR	Caregivers' concerns (interviews)
Wright, et al., 2008 (48)	332	77% 65% Caucasian, 17% African American 16% Hispanic, 2% Other	Cancer Pre and post death	Quasi-experimental, prospective	70	Caregivers' bereavement adjustment (SCID, McGill Quality of Life psychological subscale)
Wright, et al., 2010 (72)	342	75% 64% Caucasian, 20% African American 15% Hispanic, 1% Other	Cancer Pre and post death	Quasi-experimental, prospective	72	Caregivers' mental health (SCID, Prolonged Grief Disorder scale)
York, Churchman, Woodard, Wainright, & Rau-Foster, 2012 (51)	243	48% 95% Caucasian	Mostly cancer Post death	Qualitative prospective	NR	Free text comments by caregivers

^aTotal number of participants in study/number of hospice caregivers in study.

ALS, amyolateral sclerosis; BAI, Beck Anxiety Inventory; BDI, Beck Depression Inventory; BSI, Brief Symptom Inventory; CDS, Caregiver Demands Scale; CES-D, Center for the Epidemiological Studies Depression Scale; CQLI-R, Caregiver Quality of Life Revised; CQOLC, Caregiver Quality of Life Index – Cancer; CSS, Caregiver Satisfaction Scale; DSM-IV, Diagnostic and Statistical Manual, 4th ed; ENRICHD, Enhancing Recovery in Coronary Heart Disease; FAMCARE, Family Satisfaction with Advanced Cancer Care Scale; MOS; SF-36, Medical Outcome 36-Item Short-Form Health Survey; MSAS, Memorial Symptom Assessment Scale; NR, not reported; PSS, Perceived Stress Scale; SCARED, Stressful Caregiving Adult Reactions to Experiences of Dying scale; SCID, Structered Clinical Interview for DSM-IV.

Most of the included studies used non-representative samples of caregivers predominantly of cancer patients. Participants in the included studies were primarily female (with two exceptions^30,43) and Caucasian (with one exception³⁹). When reported, participant response varied significantly from a low of 22% to 94%. Few studies reported non-respondent data, making it difficult to exclude selection bias.

The nature and quality of the outcome measures used by the studies varied greatly. Only one study used a biological marker of well-being (mortality),⁴⁴ with most studies using either previously validated self-report measures or satisfaction questionnaires unique to the specific study. The Caregiver Quality of Life Index – Cancer,⁴⁵ Caregiver Quality of Life Index – Revised,⁴⁶ and the CES-D⁴⁷ were the most commonly used self-report measures. The majority of studies evaluated caregiver outcomes prior to the death of the patient. However, a few studies took a broader perspective on caregiving and assessed caregiver outcomes both before and after the death of the patient.^{30,37,48–50}

Synthesis of studies

Synthesis of the included studies showed them to fall into five main groups: (1) caregivers' satisfaction with hospice services, (2) caregivers' well-being, (3) predictors of caregivers' psychological or physical well-being, (4) the impact of hospice services on caregiver outcomes, and (5) the effectiveness of hospice-based caregiver interventions.

Caregivers' satisfaction with hospice services

These studies measured caregiver satisfaction after the death of the patient and found the majority of caregivers to be satisfied or very satisfied with hospice care and services.^{24,27,39,48,51–56} Caregivers of hospice patients reported fewer unmet needs when compared with caregivers of patients who died in traditional hospital settings.⁵⁴ In general, caregiver characteristics (i.e., ethnicity, gender, age, relationship with patient) were not found to be associated with caregiver satisfaction.^24,52 Factors associated with higher levels of caregiver satisfaction included time spent with the patient, clear goals of care, personalized treatment plans, accessibility and continuity of hospice staff, effective symptom management, and good communication between hospice staff and caregivers.^51,52,55

While level of caregiver satisfaction was not found to be associated with caregiver ethnicity, other group differences were reported between Caucasian and Latino caregivers. Caucasian caregivers access hospice services at a greater rate than Latino caregivers and the locus of decision making differs (family versus individual). Critically identified were the need for culturally specific language and improved cultural literacy among medical providers serving Latino communities.³⁹

Caregivers' well-being

Studies in this category reported on one or more aspects of caregiver well-being. Primary caregiver outcomes included mood and anxiety symptoms, complicated grief, self-reported physical health, and quality of life. Anxiety and depressive symptoms as well as caregiver stress were found to be clinically significant in caregivers of hospice patients both while actively providing care and within the first few months after the death of the patient.^{9,11,28,30,32,49,57,58} Caregiver quality of life outcomes were mixed. A few studies reported that caregivers' quality of life was relatively healthy or at least adequate,^35,59 while others reported caregivers' quality of life was lower than in noncaregiving adults.^10,46,57

Subjective indicators of physical health were found to be lower in caregivers at time of hospice enrollment compared with non-caregiving adults.⁵⁷ As the death of the patient approached, caregivers' quality of life was found to decline, and the demands of providing care substantially increased.^{10,11,35,38,59}

Several studies identified caregivers as “second order patients”⁶⁰ in need of instrumental and emotional support as well as specialized education in the areas of understanding/interpreting patient symptoms, pain management, and emotional preparation for the patient's death.^{10,37,58,60–62}

Predictors of caregivers' psychological or physical well-being

Studies in this category investigated the impact of caregiver characteristics (e.g., age, gender, health, and religiosity); patient factors (diagnosis and duration of caregiving); caregivers' cognitive appraisals; and social support on caregiver outcomes such as grief, burden, depression, and quality of life. Results for several of the predictors were equivocal. Some studies found specific patient diagnoses (i.e., neurologic disease) predictive of poorer caregiver outcomes,^53,63,64 while one study that compared two matched groups of caregivers (lung cancer and dementia) found no difference in depression and life satisfaction.⁵⁷ A few studies found female gender predictive of higher levels of depression and less life satisfaction,^34,57 while others reported caregiving daughters experienced less grief and burden than spousal caregivers.^30,53 Increased caregiver age predicted less caregiving burden and strain,^26,63 higher quality of life,⁶⁵ and less complicated bereavement.³⁰ Conversely, some studies found no significant relationship between caregivers' age, gender, and quality of life⁶⁶ or between quality of life and whether the caregiver was a spouse or daughter of the care receiver.⁶⁷

Health status, religiosity, social connections, and cognitive appraisal emerged as more reliable predictors of caregiver outcomes. Poor self-reported caregiver health status predicted higher levels of depression, less life satisfaction, and a more intensive grief response.^29,30,34,45 Greater religiosity among caregivers on the other hand predicted fewer incidents of Major Depressive Disorder 13 months after the death of the patient.⁶⁸ Limited social networks and restricted daily activities predicted increased burden, higher levels of depression, and less life satisfaction in active caregivers.^26,34,36,63 One of the most consistent findings was the importance of caregivers' subjective appraisals or perception of burden. Caregivers who appraised caregiving demands as less stressful and/or who found meaning and benefit in caregiving experienced lower levels of depression and strain and higher levels of life satisfaction.^{26,34,36,63,69}

Impact of hospice services on caregiver outcomes

This category of studies examined the impact of hospice services on caregiver mortality, psychiatric symptoms, burden, quality of life, and complicated or prolonged grief. A large retrospective study found hospice use decreased the risk of mortality for the surviving spouse, but only for women.⁴⁴ However, enrollment in hospice was not found to reduce caregiver burden,⁷⁰ to improve the quality of life for caregivers,^48,66 or to be predictive of bereavement adjustment.⁶⁷

Although hospice care was not a direct predictor of caregiver quality of life, receiving hospice services for at least a week predicted higher patient quality of life; improved patient quality of life was in turn associated with better caregiver quality of life.^48,66 Enrollment in hospice services was found to reduce the risk of Major Depressive Disorder following the patient's death⁵⁰ and also reduce the number of complicated grief symptoms.⁷¹ When comparing hospice caregivers to those in other settings, a patient's death in hospice care as opposed to traditional hospital deaths resulted in a decreased risk for Posttraumatic Stress Disorder and prolonged grief in caregivers.⁷²

Effectiveness of hospice-based caregiver interventions

This emerging group of studies developed and tested the effectiveness of caregiver interventions delivered within a hospice setting. Findings suggested brief caregiver interventions can be delivered successfully^31,41,42,73 and that technology (i.e., videophones and telephones) could improve communication between the hospice team and caregivers, thereby enabling greater participation in care decisions.^31,41,74,75

Two pilot studies tested psychosocial caregiver interventions (problem solving) delivered via videophone and telephone and failed to find significant intervention effects.^33,73 Nevertheless, a subsequent randomized control trial of a problem solving intervention offered to caregivers via videophone was found to reduce caregiver anxiety and improve problem solving skills, but had no effect on caregivers' quality of life.³¹ A comparison study of face-to-face and videophone delivery of a problem solving caregiver intervention found no significant difference in caregiver outcomes between the two methods of intervention delivery,⁴¹ suggesting a role for this technology in hospice care.

In a randomized control trial by McMillan et al., caregivers of cancer patients were found to significantly benefit (as measured by improved quality of life and reduced burden) from a brief three-session manualized psychosocial intervention.⁴⁰ This nine-day (three-visit) intervention, labeled Creativity, Optimism, Planning, and Expert Information (COPE), adapted strategies from the problem-solving research literature to address the needs and issues faced by families caring for persons with cancer. Another psychosocial intervention study used a targeted intervention to enhance the positive aspects of caregiving (meaning, purpose, and value in the caregiving experience), improving caregivers' comfort with caregiving tasks and facilitating emotional closure with the patient.⁷⁶ Results from this nine-session (7.5 hour) intervention found enhanced caregiver perceptions of the positive aspects of caregiving, comfort, and closure.

A second randomized controlled trial by McMillan et al.⁴² tested the effect of providing brief feedback from standardized psychosocial assessments of patient/caregiver dyads at interdisciplinary hospice team meetings. Results were encouraging for patients, but no change was detected in the caregiver outcomes of depression or spiritual needs. Interventions may also include tangible supports, e.g., caregiving assistance, transportation, cleaning, meal preparation. Providing these additional supports to hospice caregivers was found to reduce caregiver stress and inpatient stays when compared with hospice caregivers not receiving these supplementary services.⁷⁷

Discussion

In this review of the literature regarding informal caregiving of hospice patients, several important themes emerged. First, caregivers were found to be generally satisfied with hospice services. Clear goals of care, good communication between hospice staff and caregivers, and additional time spent with patients and families promoted higher levels of satisfaction with hospice services. Second, caregivers were found to have clinically significant levels of anxiety, depression, and perceived stress as well as poorer physical health and decreased quality of life when compared with non-caregivers. Third, when attempting to predict caregiver strain, it was found that lowered social support and worsened physical health could be related to increased depression, grief, burden, and diminished life satisfaction. Conversely, increased age, religiosity, and a self-perception about the benefits and meaning of caregiving appeared to have positive and protective effects.

Hospice care in general was not proven in these studies to reduce caregiver burden, improve caregiver quality of life, or assuage the bereavement process. However, the use of hospice services decreased the risk of mortality for female spousal caregivers and reduced the risk of depression and complicated grief after the patient's death.

Hospice-specific caregiver interventions showed significant promise. These brief structured interventions promoted problem solving and/or meaning-based coping strategies and were found to improve caregivers' quality of life, decrease burden, and enhance positive aspects of caregiving. The psychosocial interventions were offered upon admission to a hospice program and successfully delivered in-person or via videophone.

Limitations

The studies included in this review were diverse in terms of design, assessed outcomes, and sample sizes. The absence of control groups and underutilization of repeated measures or time-series designs over the course of the dying and bereavement process limited the findings of many studies. The vast majority of participants were Caucasian females caring for a person dying of cancer, which may also be problematic for overall generalizability. Finally, there was an overreliance on self-reporting and study-specific measures to assess outcomes.

Conclusions

This study highlights the invaluable role played by informal hospice caregivers and calls for systematic research on how to maximize the positive aspects of caregiving. With advances in policy, research, and resource allocation, hospices will be better able to support the critical needs of caregivers for the dying, whose vital role enables high-quality end-of-life care for so many.

Future directions

This review was helpful in synthesizing the current literature on informal caregiving in hospice; however, future research is warranted. For example, few studies adequately addressed the impact of disease trajectories on caregivers during the period of actively providing care and during the bereavement process. Future research could consider the impact different disease trajectories have on caregivers and their well-being, and how providing support to caregivers earlier in the course of their loved one's terminal illness could be beneficial. Unfortunately, the disparities apparent in the broader health care system are found in hospice care; most U.S. hospice patients are female (56%), over the age of 65 (84%), of non-Hispanic origin (93%), and Caucasian (82%).⁷⁸ Additional research is needed to better understand how to effectively support diverse caregivers not typically represented here, and to identify their specific needs once services are received.

Further research is needed to assess brief interventions that promote effective coping and enhance the positive aspects of caregiving, while also identifying factors that alleviate the detrimental aspects of caregiving. Since caregivers' cognitive appraisals of burden and stress as well as their ability to find meaning in the caring experience were predictive of enhanced well-being, future studies could more directly target and modify these subjective appraisals. Finally, as social support was associated with increased well-being and reduced burden, developing interventions that help caregivers identify and better use their informal social support networks could be beneficial.

Author Disclosure Statement

No competing financial interests exist.