Family Caregiver Reports of Their Own and Patient Symptoms in Cancer Home Hospice Approaching End-of-Life

Abstract

Context: Family caregivers assume the primary responsibility of assessing and managing hospice cancer patient symptoms while simultaneously managing their own wellbeing and symptoms. Objectives: Describe caregivers’ assessment of hospice cancer patient and their own symptoms during the last 60 days of patient life, and assess the relationship between patient and caregiver symptoms over time. Methods: Caregiver symptom report of self and cancer home hospice patient symptom data were collected via telephone in the final 60 days of patient life. Descriptive data on symptom severity and prevalence were summarized. Exploratory Factor Analysis was used to group individual symptoms. Factors representing patient symptoms, caregiver symptoms, and caregiver outlook were analyzed using mixed-effects analysis to determine relationships between factors and change in relationship between factors over time. Results: Data from 61 patient-caregiver dyads are presented. At least 1 day of moderate-to-severe symptoms were reported in the majority of dyads. Significant auto-regressive associations were found, namely previous factor scores for an individual positively predicted the next factor scores for that individual. Previous caregiver report of patient symptoms was also positively associated with the next report of caregiver symptoms; previous caregiver report of their own symptoms were negatively associated with their next report of patient symptoms. Patient and caregiver symptoms and caregiver outlook worsened over time and the relationship between patient and caregiver symptoms strengthened closer to death. Conclusion: Our findings may guide hospice care team responses to caregiver and patient symptoms to promote individual level and unit level functioning.

Keywords

oncology home hospice caregiving symptom end-of-life longitudinal

Extensive research has focused on understanding and managing symptoms, such as pain, fatigue, or distress, in patients with advanced cancer.^1-3 Research shows that, supported by an interdisciplinary care team, caregivers are involved in actively monitoring symptoms for home hospice patients and often act as a primary decision-maker regarding a course of intervention and administering care.^4,5 Patient symptoms can be dynamic, often intensifying as death approaches⁶ and caregivers report having a high number of unmet needs and report feeling unprepared for their role.⁷ Thus, the caregiving role is often associated with high levels of burden and psychological distress.⁸ Further, many caregivers are older adults and have their own health issues and related symptom experiences,⁹ which may be exacerbated by caregiving.¹⁰ Though many studies capture caregiver psychological symptoms,^11,12 such as anxiety, depression, or grief, fewer capture caregiver physical symptoms, such as sleep, fatigue, and pain.^13-15

Capturing caregivers’ report of both the patient and their own symptom data could provide additional evidence for targeted interventions. Symptom data, especially when paired with decision-support tools, can help hospice nurse care managers direct interdisciplinary care needs.^16,17 There is strong evidence patient and caregiver psychological and physical health is highly interdependent, such that patient health can impact caregiver health and vice versa.^18,19 For example, hospice cancer patient fatigue has been shown to significantly impact caregiver depression.²⁰ Similarly, the presence of caregiver chronic conditions have been shown to impact advanced cancer patient well-being.²¹ Few studies have assessed the relationship between caregivers’ report of the patient’s health and their own symptoms over time, and particularly at end of life.

The goal of this study was first to describe home caregivers’ report of hospice cancer patient and their own symptoms over the last 2 months of the patient’s life, and second, leverage repeated measures to explore the day-to-day associations between these reports over time. We hypothesize that caregiver report of their own and patient symptoms will be associated over time, particularly as death approaches. Ultimately, this information could be guidance on the complex interactions of patient and caregiver symptom burden and potential avenues to intervene.

Methods

Study Overview

Data were collected as part of a multi-site longitudinal, observational study of cancer home hospice patients and their informal family caregivers (kin or non-kin). Caregiver participants were identified through participating hospice agencies between 2017 and 2020. Consent was obtained from the caregiver and also from patients if they were physically and/or cognitively able. At the home visit, the caregivers completed a baseline survey and were asked to complete daily symptom ratings for the patient and themselves via Interactive Voice Response (IVR) telephone reporting system based on the [previous study blinded] list of patient symptoms.²² This study was approved by the Blinded Institutional Review Board (IRB_blinded).

Participants

Home hospice patients were considered eligible for the study if they were 18 years or older with a primary diagnosis of cancer, had a prognosis of 1 week or more to live, and had an eligible caregiver. Caregivers who were 18 years or older and could speak and understand English were eligible. Detailed screening and recruitment processes are detailed elsewhere.²³ The current study focused on caregivers where the patient died in hospice and only on the reports made while the patient was still alive.

Measures

Patient demographic data were recorded from hospice records. Caregivers completed baseline sociodemographic questionnaires and were asked to make daily automated reports to provide the presence, severity and level of distress of symptoms for the patient and for themselves within the last 24 hours. If a symptom was present, they were asked to rate on a 1-10 scale: (1) perception of patient symptom severity for 8 symptoms including delirium, depressed mood, diarrhea or constipation, fatigue, nervousness/anxiety, pain, poor appetite, and trouble sleeping; (2) level of distress caregiver was experiencing for each of the 8 patient symptoms; and (3) caregiver ratings of their own symptoms and outlook, including anxiety, depression, fatigue, trouble sleeping, guilt, sense of peace, and their sense of positivity. Single-item symptom reports are considered valid and reliable²⁴ and were selected to reduce burden for caregivers and because of their clinical relevance.²²

Analysis

Descriptive Statistics

Demographic variables were summarized. Symptom severity on the 0-10 scale was categorized into none (0), mild (1-3), moderate (4-7), and severe (8-10) in alignment with previous research.^16,25 Items rating peace and positivity were reverse scored to align with other ratings. Descriptive statistics of symptom severity ratings and prevalence were calculated.

Symptom Dimension Reduction

To determine how caregiver-reported outcomes could be combined into fewer meaningful domains, exploratory factor analysis (EFA) was conducted in MPlus with Varimax rotation. To allow potential for differences in factor structure as death approaches, we conducted separate EFAs for 60-41, 40-21, and 20-0 days before death. To allow for missing data, factor mean scores were computed by prorated mean of the outcomes in each factor with a requirement that ≥70% of questions be answered. The number of factors to be retained were based on scree plot and examination of eigenvalues.²⁶ Benchmarks for simple structure in the rotated loadings include loadings ≥.40 on any single factor and <.3 for dual loadings. Combining results from the three EFAs gave a rough guideline for dimension reduction. Separate Cronbach’s alpha was conducted on each of the identified factors with criteria of ≥.7 for acceptable, ≥.8 good, and ≥.9 excellent internal reliability.²⁷ Inter-factor correlations are also reported.

Modeling Caregiver Reports Over Time From Patient Death

To explore the sequential associations between outcomes over time and as death approaches, we conducted a series of mixed-effects analyses with random intercepts (random effects) about the overall population average coefficient (fixed effects). No imputation or aggregation was conducted. For each factor identified in the EFA, we constructed a lagged value (prior outcome) representing the mean for the factor at the previous call. Full models for each of the EFA factors incorporated the following potential predictors and interaction terms: (1) autoregressive-lagged (eg the degree to which last report of patient symptoms predicts current report of patient symptoms); (2) cross-lagged pathways (eg the degree to which last report’s caregiver’s own symptoms predicts current report of patient symptoms), (3) weeks till death (0 indicating day of death and negative numbers indicating time further from death), and (4) interaction terms of weeks till death with auto- and cross-lagged predictors to examine if the strength of association of lagged report with current reports changes in magnitude as death approaches. As this was an exploratory analysis, we applied a backwards stepwise procedure, first removing non-significant interaction terms with the highest P values then non-significant main effect terms. Final models report fixed effects with 95% confidence intervals, variance for the random effects, and effect sizes (estimated by dividing random effect by intercept variance).

Results

Demographics and Description of Symptom Report Reports

Sixty-one caregiver and patient dyads were included in analysis. Caregivers enrolled in the study a median of 34 (Quartile 1 = 16.5, Quartile 3 = 58) days before death, had their first and last call a median of 30 (Quartile 1 = 12, Quartile 3 = 52) and 4 (Quartile 1 = 1, Quartile 3 = 12) days before death, respectively. Demographics of both caregivers and patients are presented in Table 1. During the 60-day timeframe preceding death, caregivers reported symptoms a median of 8 days (range 1-45 days; Quartile 1 = 3 days, Quartile 3 = 16 days), with a median of 58% call compliance based on total number of reports (range 8-100%, Quartile 1 = 34%, Quartile 3 = 88%). Of 710 total reports, only those reporting on all symptoms (N = 647) were utilized for the analysis; 502 reports (71%) reported symptoms and 145 reports (20%) reported no symptoms.

Table 1.

Demographics of Caregivers and Patients.

Caregiver Social Demographics	N =61 (%)
Age
Mean ± SD	59.2 ± 13.4
Missing	2
Gender
Male	17 (27.9)
Female	44 (72.1)
Race
White/Caucasian	55 (90.2)
Asian	1 (1.6)
American Indian, Alaska Native, Aleut, Eskimo	1 (1.6)
Multiple races	4 (6.6)
Ethnicity
Non-Hispanic	54 (88.5)
Hispanic	7 (11.5)
Highest Education
High school or Equivalent or less	10 (16.4)
Some college or vocational school	21 (34.4)
College graduate	13 (21.3)
Some graduate or professional school	4 (6.6)
Graduate or professional degree	13 (21.3)
Employment
No	23 (37.7)
Part time	10 (16.4)
Full time	27 (44.3)
Missing	1 (1.6)
Perceived adequacy of financial situation
Not very good	13 (21.3)
Comfortable	34 (55.7)
More than adequate	11 (18.0)
Missing	3 (4.9)
Caregiving characteristics
Relationship to patient
Spouse/Partner	30 (49.2)
Sibling	3 (4.9)
Child	23 (37.7)
Friend/Other	5 (8.2)
Length of caregiving
<6 months	16 (26.2)
<1 year	8 (13.1)
1-2 years	10 (16.4)
3-4 years	7 (11.5)
4-5 years	6 (9.8)
6-10 years	7 (11.5)
>10 years	6 (9.8)
Missing	1 (1.6)
Co-residence with patient
Yes	46 (75.4)
No	15 (24.6)
Patient Characteristics	n=61 (%)
Age
Mean ± SD	72.70 ± 10.96
Missing	2
Gender
Male	30 (49.2)
Female	31 (50.8)
Race
White/Caucasian	58 (95.0)
Asian	1 (1.6)
Unknown	1 (1.6)
Missing	1 (1.6)
Ethnicity
Non-Hispanic	36 (59.0)
Hispanic	1 (1.6)
Missing	24 (39.3)

Symptom Prevalence

Table 2 shows prevalence, mean severity of symptoms as well as proportion of reports at moderate or severe levels. Caregivers reported fatigue (68.5%), pain (49.6%), and poor appetite (43.1%) as most prevalent patient symptoms at moderate to severe levels. For caregivers’ own symptoms, the most prevalent symptoms reported at the moderate to severe rating across all calls were lack of a sense of peace (74.2%), not feeling positive (69.7%), and fatigue (57%). The proportion of caregivers that reported their patient as having at least 1 day of moderate or severe symptoms 60 days before patient’s death, included 96.7% endorsed fatigue followed by 86.9% for both pain and poor appetite. Among the caregiver’s own symptoms at moderate or severe level, fatigue was the most prevalent symptom reported by 95.1% of the caregivers, followed by not being positive (93.4%; reverse coded) and not peaceful (93.4%; reverse coded). Though caregiver distress at patient symptom severity was also collected, bivariate analysis found high associations between distress and severity (ranging from .764 for fatigue to .938 for nervousness). As such, analysis below focuses on reports of symptom severity only.

Table 2.

Prevalence and mean severity of symptoms at moderate or severe level.

	All Calls Mean ± SD (N = 647)	# Calls Mod/Severe^a	% Calls Mod/Sev of Total Calls	# At Least 1 Call Mod/Sev (N = 61)	% at Least 1 with Mod/Sev (N = 61)
Patient symptoms
Patient fatigue	5.26 ± 3.62	443	68.5	59	96.7
Patient pain	3.36 ± 3.28	321	49.6	53	86.9
Patient poor appetite	3.29 ± 3.77	279	43.1	53	86.9
Patient delirium	2.13 ± 3.12	198	30.6	49	80.3
Patient depressed mood	1.71 ± 2.72	162	25.0	46	75.4
Patient Nervous	1.83 ± 2.93	171	26.4	44	72.1
Patient diarrhea or constipation	1.31 ± 2.68	123	19.0	42	68.9
Patient trouble sleeping	1.23 ± 2.53	119	18.4	34	55.7
Caregiver symptoms
Caregiver peace^*	5.58 ± 2.53	480	74.2	57	93.4
Caregiver positive^*	5.17 ± 2.40	451	69.7	57	93.4
Caregiver fatigue	4.56 ± 2.54	369	57.0	58	95.1
Caregiver anxiety	4.00 ± 2.65	333	51.5	56	91.8
Caregiver trouble sleeping	3.68 ± 2.49	288	44.5	51	83.6
Caregiver depress	3.31 ± 2.55	256	39.6	53	86.9
Caregiver guilt	1.92 ± 2.61	131	20.2	38	62.3

Note: * indicates the scores are reversed with higher means representing worse outcome

^aScores for moderate or severe symptoms are any symptom having ≥4 out of 0-10.

Exploratory Factor Analysis of Caregiver Report of Patient and Their Own Symptoms

Symptom report data were examined for dimension reduction using exploratory factor analysis (EFA) for the 8 patient symptoms and 7 caregiver symptoms. Table 3 provides means and standard deviations for each question, factor structure and loadings, and correlation table. Results were consistent across time increments and EFA using all reports 60 days before patient death (N=647 reports); 3 factors were optimal. Factor 1 was composed of caregiver report of Patient Symptoms, Factor 2 Caregiver Symptoms, and Factor 3 included how often caregiver felt positive or sense of peace—or “Caregiver Outlook.” Severity of patient appetite and patient nervousness had moderate loadings on factor 2 and caregiver anxiety and depression had moderate loadings on factor 1. Cronbach’s alpha for Patient Symptoms was .799, Caregiver Symptoms was .856 and Caregiver Outlook was .811. Mean scores were computed for each of the factors with the requirement of 70% or greater completed items for the factor. Accordingly, out of 647 reports, factor sums were computed for Patient Symptom (N = 641), Caregiver Symptoms (N =631), Caregiver Outlook (N = 629).

Table 3.

Exploratory Factor Analysis for symptom outcomes 60 days before patient’s death.

Outcomes (0-10 Scale)			Factors
	Mean	SD	PT sx	CG sx	CG outlook
Patient pain	3.34	3.285	.644
Patient delirium	2.14	3.131	.546
Patient bowel	1.31	2.676	.392
Patient fatigue	5.27	3.616	.694
Patient mood	1.72	2.733	.571
Patient nervous	1.82	2.928	.557	.31
Patient trouble sleeping	1.23	2.536	.392
Patient appetite	3.28	3.772	.549	.316
Caregiver fatigue	4.54	2.531		.694
Caregiver trouble sleep	3.67	2.488		.586
Caregiver anxiety	3.99	2.64	.323	.716
Caregiver depression	3.31	2.551	.308	.769
Caregiver guilt	1.92	2.611		.623
Caregiver positive*	5.17	2.4			.735
Caregiver peace*	5.58	2.53			.836
			PT sx	CG sx	CG outlook
	PT sx		.799^a
	CG sx		.507	.856^a
	CG outlook		.169	.406	.811^a

Patient Symptoms (PT Sx); Caregiver Symptoms (CG Sx); Cg Outlook (Cg outlook) *Reverse coded such that higher scores are worse ^a Cronbach’s Alpha of mean scores.

Sequential Relationships Among Caregiver Reports on Patient and Caregiver Symptoms

We conducted a series mixed-effects analysis with the three factors to explore potential sequential associations. Specifically, we investigated how the caregivers’ previous report of patient symptoms, caregiver symptoms, and caregiver outlook impacted the current report. Table 4 and Figure 1 summarize the three models.

Table 4.

Sequential relationships of previous factors patient symptoms, caregiver symptoms, and caregiver outlook on current factors including the impact of associations as death approaches.

	Patient symptoms		Effect	Caregiver symptoms		Effect	Caregiver outlook		Effect
Predictor Variable	Best Model	95% CI	Size	Best Model	95% CI	Size	Best Model	95% CI	Size
Intercept	3.10**	[2.38, 3.82]		3.59**	[2.96, 4.22]		3.66**	[3.08, 4.23]
Weeks till death (time)	.28**	[.15, .41]	.24	.08^a	[.00, .15]	.05	.09**	[.03, .15]	.08
Previous CG-Outlook	n.s			n.s			.39**	[.31, .47]	.35
Previous PT-Symptoms	.40**	[.26, .54]	.35	.12^a	[.01, .22]	.08	n.s
Previous CG-Symptoms	−.20^a	[-.36,-.04]	.17	.20**	[.11,0.29]	.13	n.s
time^aprevious PT-sx	.05**	[.01, .08]	.04	.04**	[.02, .07]	.03	n.s
time^aprevious CG-sx	−.06**	[-.10, −.02]	.05	n.s			n.s
time^aprevious CG-out	n.s			n.s			n.s
Error (var)	1.86**			1.12**			1.72**
Intercept (var)	1.34**			2.24**			1.24**

^aP <.05, **P <.01.

PT-Sx (patient symptoms); CG-Sx (caregiver symptoms); CG-Out (caregiver outlook).

Figure 1.

Lagged analysis results. Note: Each arrow defines the average fixed effect regression coefficient from the previous call outcome on current call outcome. Arrows from like outcomes are the autoregressive pathways (eg CP symptoms->CP symptoms) and those between different symptoms are the cross-lagged pathways (eg CP Symptoms ->Patient Symptoms). Because higher values represent worsening of the domain, significant positive coefficient is interpreted as previous call outcome leads to worsening or increase in current call outcome (solid arrows) while negative coefficients are interpreted as previous call outcome leading to improvement or decrease in current call outcome (long dashed arrows). Short dashed arrows represent non-significant (n.s.) associations. Cg: caregiver; pt: patient; n.s. = not significant.

As described above, models are composed of auto-regressive and cross-lagged predictors. There were significant autoregressive associations: previous caregiver symptoms positively predicted current caregiver symptoms (.2 95% confidence interval [.11, .29]), previous patient symptoms positively predicted current patient symptoms (.4 [.25, .54]), and caregiver outlook positively predicted current caregiver outlook (.39 [.31, .47]). For cross-lagged associations, caregiver symptoms and patient symptoms had significant cross-lagged associations, such that for patient symptoms, a one-unit change in previous caregiver symptoms results in a .20 decrease in patient mean current symptoms. For caregiver symptoms, a one-unit change in previous patient symptoms results in a .12 increase in caregiver mean current symptoms. There were no significant cross-lagged associations with caregiver outlook.

For all three outcomes, for each week approaching death there was significant worsening in severity with .28 increase for patient symptoms (P < .01), .08 increase for caregiver symptoms (P < .05), and .09 for caregiver outlook (P < .01). There were also significant interaction terms of the relevant past lagged predictors with death in weeks, such that (1) the magnitude of the impact of previous patient symptoms on current patient symptoms increased with proximity to week of death and (2) the impact of previous caregiver symptom on current patient symptom severity was stronger with proximity to death. For caregiver symptom severity, as patient symptom severity increased, there was a stronger negative impact on caregiver symptom severity as death approaches. There were no significant interaction effects for caregiver outlook (see Table 4).

Finally, there was significant variation for an individual’s overall symptom severity from the population average as reflected by the random intercept variance for all three outcomes. Variance for an individual’s predicted single factor score were also high for all three outcomes. These results suggest that there is unexplained between- and within-individual variance that is not accounted for in the current analysis.

Discussion

This study is one of the first to assess symptoms based on family caregiver report of both cancer home hospice patients and their own symptoms across time. In line with previous research primarily focused on cancer patient symptoms,¹⁶ our study finds high proportion of reports of moderate-to-severe fatigue and pain, suggesting there is work to be done to improve symptom management at end of life in hospice. Although a limitation is that our patient symptom reports are made by caregivers, in hospice, daily patient care and hospice team response is often based on caregiver report. Even if caregivers are over-estimating patient symptoms, our findings suggest that this misperception still has important impacts on caregiver symptoms and outlook. Additionally, high levels of variability in all symptom reports exist both within and between individuals. While variability is expected due to the dynamic nature of hospice, systems are needed to help anticipate trends and alert hospice staff to address worsening symptoms. New technology can efficiently capture these changes and alert staff when visits should be intensified.²²

Few studies have specifically assessed daily reports of caregiver symptoms; our findings contribute to previous work noting hospice family caregiver physical and mental health are at risk.^28,29 In addition to more physical symptoms, we also assess caregiver outlook variables of positivity and sense of peace. Growing research has identified positive aspects of caregiving, such as a sense of personal satisfaction, as protective for family caregivers.³⁰ Similarly, characteristics of spirituality, including peace, have been shown to be important aspect of caregiving and coping.^8,31 The low levels of positive outlook in our study suggests there may be a need to better address these aspects in family caregivers. More research is needed to determine how caregiver outlook constructs such as peace and positivity interact with other more psychological and spiritual constructs, as well as the role they play in hospice caregiving.

A growing literature has demonstrated the relatedness of patient and caregiver outcomes.^18,19 However, less work has assessed their interdependence over time. As symptoms can be dynamic, especially as death approaches, this is an important feature of the current study. As expected, the association between sequential ratings of patient symptoms strengthened approaching death. While the change over each report is small, it is statistically significant and because we asked for daily reports, this means that clinically-meaningful changes can develop over a relatively short amount of time. This finding reinforces the goal of addressing symptoms early--often it is more difficult to manage severe symptoms than minor symptoms—and provides additional evidence for the benefit of frequent symptom assessment.^16,17

We also find the caregiver’s report of patient’s past symptoms is positively related to the caregiver’s own next symptoms, such that when the caregiver reports worse patient symptoms, the caregiver then reports worse symptoms themselves on their next rating. However, the caregiver’s past symptoms are negatively related to the patient’s next symptoms, such that when the caregiver reports worse symptoms for themselves, the caregiver actually reports better symptoms for the patient on the next rating. As death approached, these relationships between caregiver perceptions of the patient’s and their own symptoms strengthened, perhaps due to the increased need for care. One explanation for the somewhat counterintuitive negative relationship between caregiver symptoms and subsequent patient symptoms is that as caregivers work hard to improve patient symptoms, their perception of their own health and well-being suffers. Similarly, when caregivers believe patients’ symptoms are improving, this may produce complex emotions for caregivers; while glad for the improvement, some caregivers may view the improvement as a sign that the hard work of caregiving will be further extended.³² Caregiving is often perceived as being a self-sacrificial act,³³ and evidence suggests caregivers are making this sacrifice without adequate resources and support.^34,35 These exploratory and hypothesis-generating findings merit future studies. They also support continued research on promoting respite for caregivers and other psychosocial interventions.^36,37

Limitations

This study is one of few to gather daily reports from caregivers of both patient and their own symptoms during home hospice care. Although patient reports can be difficult to obtain at end-of-life, there may be some bias in reporting based on caregiver perception.³⁸ More detailed standardized and validated questionnaires may also have captured greater precision in symptom reports. Our study also had missing data, and this may particularly impact our lagged analysis. Specifically, the sequential lagged analysis assesses the impact of the previous report on the current report which with missing data may not be based on reports on consecutive days. As such, our results likely underestimate the impact of perceived patient and caregiver symptom burden on current symptom reports. Furthermore, as this study was based on end of life and hospice care, there was heterogeneity in the timing of when participants joined the study prior to patient death. Thus, some caregivers had fewer opportunities to report on symptoms and consequently provided less information related to change in associations as the patient nears death. More complete datasets are needed to better understand the inter-relationships of caregiver reports of patient and caregiver symptoms. Further, additional symptoms not assessed in this study may have been important to patients and caregivers. A larger sample would also allow an examination of differences in symptom reporting by gender, race/ethnicity, and patient-caregiver relationship.

Conclusion

Symptom management continues to be an issue for cancer home hospice patients and their family caregivers. More policies, clinical services and interventions should be developed for family caregivers to better support their own health and protect their ability to care for the patient, particularly as they play a key role in hospice care.

Footnotes

Authors Contributions

Maija Reblin, Eli Iacob and Gary Donaldson conceptualized and designed the manuscript analysis. Lee Ellington was awarded funding to support the work. Maija Reblin and Eli Iacob wrote the first draft of the manuscript. Djin L. Tay, Megan C. Thomas Hebdon, Anna Beck, Kristin G. Cloyes and Lee Ellington contributed to later drafts. Eli Iacob, Hui Li and Gary Donaldson conducted analyses. All authors read and approved the final manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Institute of Nursing Research (Grant number R01 NR016249). Additional author time was supported by the National Institute of Nursing Research (T32 NR013456; MCTH) and the National Institute on Aging (K07 AG068185; LE).

Ethical Approval

This study was approved by the University of Utah Institutional Review Board (IRB_00033122). The authors certify that the study was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.

Informed Consent

Informed consent was obtained from all individual participants included in the study.

Data Availability

Data is available upon reasonable request from the corresponding author.

ORCID iD

Maija Reblin

References

Teunissen

SCCM

Wesker

Kruitwagen

de Haes

HCJM

Voest

de Graeff

. Symptom prevalence in patients with incurable cancer: a systematic review. J Pain Symptom Manag. 2007;34(1):94-104. doi:10.1016/j.jpainsymman.2006.10.015

Kwekkeboom

. Cancer symptom cluster management. Semin Oncol Nurs. 2016;32(4):373-382. doi:10.1016/j.soncn.2016.08.004

Stapleton

Holden

Epstein

Wilkie

. Symptom clusters in patients with cancer in the hospice/palliative care setting. Support Care Cancer. 2016;24(9):3863-3871. doi:10.1007/s00520-016-3210-6

Ullgren

Tsitsi

Papastavrou

Charalambous

. How family caregivers of cancer patients manage symptoms at home: a systematic review. Int J Nurs Stud. 2018;85:68-79. doi:10.1016/j.ijnurstu.2018.05.004

Thana

Lehto

Sikorskii

Wyatt

. Informal caregiver burden for solid tumour cancer patients: a review and future directions. Psychol Health. 36, 2021:1-23. doi:10.1080/08870446.2020.1867136

Verkissen

Hjermstad

Van Belle

Kaasa

Deliens

Pardon

. Quality of life and symptom intensity over time in people with cancer receiving palliative care: results from the international European Palliative Care Cancer Symptom study. PLoS One. 2019;14(10):e0222988. doi:10.1371/journal.pone.0222988

Oechsle

. Current advances in palliative & hospice care: problems and needs of relatives and family caregivers during palliative and hospice care—an overview of current literature. Med Sci. 2019;7(3):43.

Ochoa

Buchanan Lunsford

Lee Smith

. Impact of informal cancer caregiving across the cancer experience: a systematic literature review of quality of life. Palliat Support Care. 2020;18(2):220-240. doi:10.1017/S1478951519000622

Sambasivam

Liu

Vaingankar

, et al. The hidden patient: chronic physical morbidity, psychological distress, and quality of life in caregivers of older adults. Psychogeriatrics. 2019;19(1):65-72. doi:10.1111/psyg.12365

10.

Shaffer

Kim

Carver

Cannady

. Effects of caregiving status and changes in depressive symptoms on development of physical morbidity among long-term cancer caregivers. Health Psychol. 2017;36(8):770-778. doi:10.1037/hea0000528

11.

Adelman

Tmanova

Delgado

Dion

Lachs

. Caregiver burden: a clinical review. JAMA : The Journal of the American Medical Association. 2014;311(10):1052-1060.

12.

Geng

H-M

Chuang

D-M

Yang

, et al. Prevalence and determinants of depression in caregivers of cancer patients: a systematic review and meta-analysis. Medicine (Baltimore). 2018;97(39):e11863. doi:10.1097/MD.0000000000011863

13.

Washington

Parker Oliver

Smith

McCrae

Balchandani

Demiris

. Sleep problems, anxiety, and global self-rated health among hospice family caregivers. Am J Hospice Palliat Med. 2018;35(2):244-249. doi:10.1177/1049909117703643

14.

Oechsle

Ullrich

Marx

, et al. Prevalence and predictors of distress, anxiety, depression, and quality of life in bereaved family caregivers of patients with advanced cancer. Am J Hospice Palliat Med. 2020;37(3):201-213. doi:10.1177/1049909119872755

15.

Fletcher

BAS

Dodd

Schumacher

Christine Miaskowski

. Symptom experience of family caregivers of patients with cancer. Oncology Nursing Society; 2008;35(2):E23-E44.

16.

Mooney

Beck

Wong

, et al. Automated home monitoring and management of patient-reported symptoms during chemotherapy: results of the symptom care at home RCT. Cancer Med. 2017;6(3):537-546. doi:10.1002/cam4.1002

17.

Mooney

Iacob

Wilson

Lloyd

Nielson

Ward

. Randomized trial of remote cancer symptom monitoring during COVID-19: impact on symptoms, QoL, and unplanned health care utilization. J Clin Oncol. 2021;39(15_suppl l):12000-12000. doi:10.1200/JCO.2021.39.15_suppl.12000

18.

Streck

Wardell

LoBiondo-Wood

Beauchamp

JES

. Interdependence of physical and psychological morbidity among patients with cancer and family caregivers: review of the literature. Psycho Oncol 2020;29(6):974-989. doi:10.1002/pon.5382

19.

Kershaw

Ellis

Yoon

Schafenacker

Katapodi

Northouse

. The interdependence of advanced cancer patients’ and their family caregivers’ mental health, physical health, and self-efficacy over time. Ann Behav Med. 2015;49(6):901-911. doi:10.1007/s12160-015-9743-y

20.

Buck

Benitez

Fradley

, et al. Examining the relationship between patient fatigue-related symptom clusters and carer depressive symptoms in advanced cancer dyads: a secondary analysis of a large hospice data set. Cancer Nurs. 2020;43(6):498-505.

21.

Ketcher

Otto

Vadaparampil

Heyman

Ellington

Reblin

. The psychosocial impact of spouse-caregiver chronic health conditions and personal history of cancer on well-being in patients with advanced cancer and their caregivers. J Pain Symptom Manag. 2020;62:303-311.

22.

Mooney

Whisenant

Beck

. Symptom care at home: a comprehensive and pragmatic PRO system approach to improve cancer symptom care. Med Care. 2019;57(suppl 5 1):S66-S72.

23.

Tay

Iacob

Reblin

, et al.

What contextual factors account for anxiety and depressed mood in hospice family caregivers?

Psycho Oncol. 2021;31:316-325.

24.

Cleeland

Fisch

Dunn

. Cancer Symptom Science: Measurement, Mechanisms, and Management. Cambridge University Press; 2010.

25.

Mooney

Beck

Friedman

Farzanfar

Wong

. Automated monitoring of symptoms during ambulatory chemotherapy and oncology providers' use of the information: a randomized controlled clinical trial. Support Care Cancer: Official Journal of the Multinational Association of Supportive Care in Cancer. 2014;22(9):2343-2350. doi:10.1007/s00520-014-2216-1

26.

Pett

Lackey

Sullivan

. Making Sense of Factor Analysis: The Use of Factor Analysis for Instrument Development in Health Care Research. Sage; 2003.

27.

Cronbach

. Coefficient alpha and the internal structure of tests. Psychometrika. 1951;16(3):297-334.

28.

Williams

A-L

McCorkle

. Cancer family caregivers during the palliative, hospice, and bereavement phases: a review of the descriptive psychosocial literature. Palliat Support Care. 2011;9(3):315-325. doi:10.1017/S1478951511000265

29.

Haley

LaMonde

Han

Narramore

Schonwetter

. Family caregiving in hospice: effects on psychological and health functioning among spousal caregivers of hospice patients with lung cancer or dementia. Hospice J. 2000;15(4):1-18. doi:10.1080/0742-969X.2000.11882959

30.

Loke

. The positive aspects of caregiving for cancer patients: a critical review of the literature and directions for future research. Psycho Oncol 2013;22(11):2399-2407. doi:10.1002/pon.3311

31.

Lalani

Duggleby

Olson

. Spirituality among family caregivers in palliative care: an integrative literature review. Int J Palliat Nurs. 2018;24(2):80-91. doi:10.12968/ijpn.2018.24.2.80

32.

Sand

Strang

Existential loneliness in a palliative home care setting. J Palliat Med. 2006;9(6):1376-1387. doi:10.1089/jpm.2006.9.1376

33.

Tranberg

Andersson

Nilbert

Rasmussen

. Co-afflicted but invisible: a qualitative study of perceptions among informal caregivers in cancer care. J Health Psychol. 2021;26(11):1850-1859. doi:10.1177/1359105319890407

34.

AARP and the National Alliance for Caregiving . Caregiving in the US. 2020. https://www.aarp.org/content/dam/aarp/ppi/2020/05/full-report-caregiving-in-the-united-states

35.

Family Caregiver Alliance . Caregiver Statistics: Demographics. 2016. https://www.caregiver.org/resource/caregiver-statistics-demographics/

36.

Ahn

Romo

Campbell

. A systematic review of interventions for family caregivers who care for patients with advanced cancer at home. Patient Educ Counsel. 2020;103(8):1518-1530. doi:10.1016/j.pec.2020.03.012

37.

Jadalla

Page

Ginex

Coleman

Vrabel

Bevans

. Family Caregiver Strain and Burden: a systematic review of evidence-based interventions when caring for patients with cancer. Article. Clin J Oncol Nurs. 2020;24(1):31-50. doi:10.1188/20.CJON.31-50

38.

Silveira

Given

Rosland

Piette

. Patient-caregiver concordance in symptom assessment and improvement in outcomes for patients undergoing cancer chemotherapy. Chron Illness. 2010;6(1):46-56.