Relationships among Symptom Management Burden,Coping Responses,and Caregiver Psychological Distress at End of Life

Abstract

Background:

Family caregivers (FCGs) face numerous stressors and are at heightened risk of psychological distress. While theoretical explanations exist linking caregiving stressors with outcomes such as anxiety and depression, limited testing of these theories has occurred among FCGs of patients nearing the end of life.

Objective:

Researchers sought to evaluate mediational relationships among burden experienced by hospice FCGs because of symptom management demands, caregivers' coping responses, and caregivers' psychological distress.

Design:

Quantitative data for this descriptive exploratory study were collected through survey. Hypothesized relationships among caregiver variables were examined with structural equation modeling.

Setting/Subjects:

Respondents were FCGs (N = 228) of hospice patients receiving services from a large, non-profit community hospice in the Mid-Southern United States.

Results:

Burden associated with managing hospice patients' psychological symptoms was shown to predict psychological distress for FCGs. Caregivers' use of escape-avoidance coping responses mediated this relationship.

Conclusions:

Results suggest that FCGs would benefit from additional tools to address patients' psychological symptoms at end of life. When faced with psychological symptom management burden, caregivers need a range of coping skills as alternatives to escape-avoidance coping.

Family caregivers (FCGs) of hospice patients are at heightened risk for psychological distress, such as anxiety and depression.^1–3 While the impending death of a family member or close friend is often inherently sad and worrisome, a number of caregiving stressors have been found to increase distress. Among these is the burden associated with managing the complex symptomatology often present at the end of life. FCGs may experience pronounced discomfort managing physical pain, particularly if the pain management regimen involves opioid use.⁴ Symptoms such as dyspnea and nausea are frequently difficult to manage at the end of life, due, in part, to the challenges faced by FCGs in administering prescribed treatments.⁵ In addition to physical concerns, patients' psychological symptoms (e.g., sadness, worrying, and irritability) can be especially distressful for family members and friends.⁶

FCGs' coping responses have been identified as potential mediators in the relationship between caregiving stressors and caregiving outcomes.⁷ For example, positive reappraisal (e.g., deriving a sense of self-esteem from caregiving activities) has been found to be associated with greater life satisfaction among hospice FCGs.⁸ Furthermore, testing of interventions that promote specific coping responses, such as planful problem-solving, have been shown to reduce FCGs' psychological distress and improve their quality of life.^9,10

While research on stress and caregiving is plentiful, a number of opportunities exist to strengthen the knowledge base. To date, much of the caregiving research has adopted a broad lens in its consideration of stressors, emphasizing the complex and interrelated nature of caregiving challenges, demands, and realities. Information on how FCGs cope with specific problems (e.g., symptom management demands) is limited, and the results of targeted coping strategies are largely unknown, particularly at the end of life. In addition, while hospice and palliative medicine espouse an explicitly holistic model of care,¹¹ studies of how FCGs engage in and are affected by symptom management largely focus on physical symptoms, despite the fact that FCGs find managing psychological symptoms to be more distressful.⁶ Research on how FCGs cope with specific versus generalized challenges is needed, and the field's understanding of symptom management must broaden to include management of both physical and psychological symptoms.

Study Aim and Hypotheses

In this study, researchers sought to evaluate mediational relationships among hospice FCGs' symptom management burden, coping responses, and psychological distress. The following hypotheses were tested: (1) FCGs' symptom management burden predicts FCGs' psychological distress (i.e., symptoms of anxiety and depression), (2) FCGs' coping responses predict FCGs' psychological distress, (3) FCGs' symptom management burden predicts FCGs' coping responses, and (4) the relationship between FCGs' symptom management burden and FCGs' psychological distress is mediated by their coping responses. By focusing on coping with a specific caregiving problem (symptom management burden) and separately considering management of physical and psychological symptoms, researchers aimed to enrich the field's understanding of the caregiving experience and provide actionable results to strengthen FCG support in hospice care.

Research Design

Researchers conducted a descriptive exploratory study, which was approved by the University of Louisville Institutional Review Board. In partnership with a large, community-based, nonprofit hospice agency in the Mid-Southern United States, researchers recruited a volunteer sample of FCGs to participate in a survey. Inclusion criteria required that participants be aged 18 or older, the primary caregiver of an adult hospice patient, and able to communicate in spoken or written English. During their routine visits, the hospice agency social workers asked potential participants for permission to share their contact information with the research team. If permission was granted, a researcher contacted the caregiver to explain the study, confirm study eligibility, and, for those interested in participating, make the necessary arrangements. Participating FCGs were given the option of responding to the survey online, by telephone, or in writing through U.S. mail.

Study sample

A total of 248 participants were recruited. Twenty respondents had missing values on most or all study variables; these cases were not included in the analysis. The final analytic sample consisted of 228 hospice FCGs (M age = 61.4 years, SD = 12.7), 44 of whom were male. The average length of time patients had received hospice services before their caregiver's participation in the study was 103 days (SD = 185 days). Additional demographic information for the study sample is summarized in Table 1.

Table 1.

Caregiver Characteristics and Descriptive Statistics

	n	M (SD)
Demographic variables
Age	228	61.35 (12.65)
Percent male	226	19.5%
Days in hospice	226	102.59 (184.98)
Criterion variables
MSAS physical symptom burden	207	9.33 (7.73)
MSAS psychological symptom burden	216	6.94 (5.55)
Mediator variables
Accepting responsibility	213	4.67 (1.99)
Confrontive coping	212	5.88 (2.30)
Distancing	216	5.75 (2.04)
Escape-avoidance	216	4.99 (2.11)
Planful problem-solving	212	7.25 (2.42)
Positive reappraisal	215	6.10 (2.47)
Seeking social support	219	7.05 (2.75)
Self-controlling	217	6.24 (2.11)
Outcome variables
PHQ-9	224	6.49 (5.60)
GAD-7	221	8.14 (6.06)

All mediator variables are the Ways of Coping subscale total scores.

GAD-7, Generalized Anxiety Disorder-7 total score; M, mean; MSAS physical symptom burden, MSAS physical symptom subscale total score; MSAS psychological symptom burden, MSAS psychological symptom subscale total score; PHQ-9, Patient Health Questionnaire-9 total score; SD, standard deviation.

Measures

Symptom management burden

The Memorial Symptom Assessment Scale–Short Form (MSAS-SF)¹² was the primary measure of symptom management burden. The instrument was originally intended to measure the distress associated with 32 symptoms commonly experienced in advanced disease; these symptoms have been found to load on two subscales: physical symptoms and psychological symptoms. Cronbach's alpha coefficients of internal consistency for the instrument's subscales have been shown to range from 0.76 to 0.87.¹² In addition, these subscales have demonstrated convergent validity with existing measures of functional status among individuals with advanced disease.¹² Although the MSAS-SF was originally administered to patients, modified versions have been successfully used to better understand FCGs' perceptions of patients' symptom experiences¹³ and measure the degree to which symptoms are burdensome to FCGs.¹⁰ This modified version of the MSAS-SF was used in this study. Specifically, FCGs were asked to indicate which of 32 symptoms associated with advanced disease had been experienced by their patient in the prior week and the degree to which managing those symptoms was burdensome to them as FCGs.

Caregiver coping responses

Caregiver coping responses were measured by the Ways of Coping Scale (WOCS).¹⁴ The WOCS lists specific coping responses and instructs respondents to indicate the degree, on a 4-point Likert scale (0, does not apply or not used to 3, used a great deal), to which they employed each particular response when faced with a recent stressor (for this study, FCGs were explicitly instructed to choose a stressor related to symptom management). WOCS's eight subscales consist of confrontive coping, distancing, self-controlling, seeking social support, accepting responsibility, escape-avoidance, planful problem-solving, and positive reappraisal.¹⁴ Internal consistency coefficients reported from initial validation studies for each of the eight subscales have ranged from relatively poor (α = 0.61, distancing) to acceptable (α = 0.79, positive reappraisal).¹⁴ The WOCS has, however, demonstrated construct validity through multiple studies showing its measurements to be consistent with theoretical predictions.^15–18 A confirmatory factor analysis (CFA) was previously conducted on the WOCS to determine the structure of coping responses among hospice FCGs.¹⁹ That study found that several items on each scale had statistically nonsignificant factor loadings, and their removal from the model resulted in stronger fit indices. This abbreviated version of the WOCS, which retained the three items with the strongest factor loadings for each subscale, was used in this study. The overall Cronbach's alpha for all 24 items was in the excellent range, α = 0.849.

Caregiver anxiety

The GAD-7²⁰ was used to measure the frequency with which respondents experienced symptoms of anxiety. GAD-7 total scores range from 0 to 21; higher scores reflect greater anxiety. Total scores of 5, 10, and 15 have been established as thresholds for mild, moderate, and severe anxiety, respectively. Previous research has found the GAD-7's internal consistency to be excellent (Cronbach's alpha = 0.92) and its test-retest reliability to be very good (intraclass correlation = 0.83).²⁰

Caregiver depression

The PHQ-9²¹ was used to measure the frequency with which respondents experienced symptoms of depression. PHQ-9 total scores range from 0 to 27; higher scores reflect greater depression. Total scores of 5, 10, 15, and 20 have been established as thresholds for mild, moderate, moderately severe, and severe depression, respectively. Research has determined that the PHQ-9's internal consistency and test-retest reliability are excellent (Chronbach's alpha >0.85; intraclass correlation = 0.84).²¹

Statistical analysis

The hypothesized relationships in our mediator model, displayed in Figure 1, were examined with structural equation modeling (SEM) using Mplus, Version 7.31.²² An array of fit indices was used to enhance the validity of statistical conclusions: chi-square for nested models²³; comparative fit index (CFI)^23,24; and root mean square error of approximation (RMSEA).^25,26 Overall model fit was assessed using both CFI and RMSEA, taking into account model complexity. In addition to chi-square tests of the statistical significance of improvements in fit, CFI and RMSEA indices were used to assess any practical improvements in fit. As shown in Figure 1, each group of factors (i.e., the predictor, mediator, and criterion factors) was also given a conceptual label in italics. For instance, the predictor factors of Physical Symptoms and Psychological Symptoms were conceptually identified as Patient Symptom Management Burden. Whenever possible, the conceptual labels are used for simplicity in the description of data analysis and results.

FIG. 1.

Hypothesized path diagram of Coping Responses mediating the effect of Patient Symptom Management Burden on Caregiver Psychological Distress. Thin lines represent “a” pathways. Dashed lines represent “b” pathways. Thick lines represent “c” pathways. To conserve space, observed variables for each latent factor are not shown. The shading of the Self-Controlling coping response depicts its deletion from the final model (Supplementary Materials).

SEM model specifications

This study involved two broad levels of analysis: (1) CFAs and (2) evaluation of direct or indirect effects. All analyses used Maximum Likelihood, which uses all available data and, thus, is advantageous compared to more traditional missing data approaches.²⁷

Confirmatory factor analyses

As previously mentioned, groups of factors constituted the predictors, mediators, and criterions; therefore, CFAs were a necessary first step for two reasons: one, to examine the structural validity of the hypothesized factors, and two, to evaluate the overall model fit when multiple factors were involved, before including any predicted pathways (i.e., direct and indirect effects). A single CFA that included multiple factors was considered the more parsimonious model for comparisons against the structural models; additionally, the observed variables across factors may covary. After the best-fitting factor structure was identified in any given CFA, the predicted pathways were subsequently added. The factor structure of Coping Responses (Fig. 1) was previously analyzed within this dataset.¹⁹

Direct and indirect effects

Three separate SEM models were analyzed to test predicted relationships. First, the Caregiver Psychological Distress factors were regressed on Patient Symptom Management Burden factors (the “c” pathways in Fig. 1). Second, Caregiver Psychological Distress factors were regressed on Coping Responses (the “b” pathways in Fig. 1). Third, Coping Responses were regressed on Patient Symptom Management Burden (the “a” pathways in Fig. 1). Last, a single SEM model was conducted to examine the presence of an indirect (or “mediated”) effect of Coping Responses. Only the statistically significant direct effects, identified in the previous stage of analysis, were included.

Results

Descriptive information

Mean caregiver depressive symptoms were in the mild range in terms of overall severity (PHQ-9 total M = 6.49, SD = 5.60), as were caregiver anxious symptoms (GAD-7 total M = 8.14, SD = 6.06). Additional descriptive statistics for the criterion, mediator, and outcomes variables are included in Table 1.

Relationships between patient symptom management burden and caregiver psychological distress

Following the conceptual model in Figure 1, a CFA was conducted with the two Patient Symptom Management Burden factors (Physical Symptoms and Psychological Symptoms) and the two Caregiver Psychological Distress factors (Depressive Symptoms and Anxious Symptoms). Results indicated that the best-fitting model included four first-order factors (Physical Symptoms, Psychological Symptoms, Depressive Symptoms, and Anxious Symptoms) with 10 individual covariances (Model 11D in Table 2). This model was subsequently used to test direct effects. Additional detail on model development and corresponding fit indices is provided in Supplementary Materials (Supplementary Data are available online at www.liebertpub.com/jpm).

Table 2.

Selected Model Fit Indices

Model	Model description	DF	χ²	CFI	RMSEA (90% CI)
11D	Final model for direct relationships of Caregiver Psychological Distress on Patient Symptom Management Burden factors	513	775.55	0.91	0.047 (0.04–0.054)
12C	Final model for direct relationships of Caregiver Psychological Distress on Escape-Avoidance	597	884.24	0.906	0.046 (0.039–0.052)
19	Final model for direct relationships of Coping Responses (i.e., Escape-Avoidance and Accepting Responsibility) on Patient Symptom Management Burden factors	563	823.44	0.869	0.045 (0.038–0.052)
22	Final model for indirect relationship of Escape-Avoidance on Psychological Symptoms and Caregiver Psychological Distress factors	1230	1769.03	0.869	0.044 (0.039–0.048)

Additional detail on model development and corresponding fit indices is provided in Supplementary Materials.

90% CI, 90% confidence interval for RMSEA estimate; CFI, comparative fit index; CI, Confidence Interval; DF, degrees of freedom; RMSEA, root mean square error of approximation.

Results in Table 3 include the parameter estimates of all the effects tested; the header row labels the individual models, and fit indices for final models are provided in Table 2. The final model for this stage in the analysis (Model 11D) included two direct effects: Psychological Symptoms significantly predicted Depressive Symptoms (β = 0.54, p < 0.001) and Anxious Symptoms (β = 0.58, p < 0.001). Thus, the first hypothesis was partially supported. FCGs' burden associated with managing patients' psychological symptoms significantly predicted FCGs' psychological distress, including both depressive and anxious symptoms. However, FCGs' burden associated with managing patients' physical symptoms did not directly influence FCGs' distress.

Table 3.

Parameter Estimates of Direct and Indirect Effects

Model parameter	Model 11B β	Model 11C β	Model 11D β	Model 12B β	Model 12C β	Model 17B β	Model 18B β	Model 19 β	Model 20 β	Model 22 β
PHYS predicts DEP	0.22	—	—
PHYS predicts ANX	0.28^*	0.14	—
PSYC predicts DEP	0.35^*	0.54^***	0.54^***						0.37^**	0.31^**
PSYC predicts ANX	0.33^*	0.46^***	0.58^***						0.46^***	0.36^***
ESCAPE predicts DEP				0.65^***	0.63^***				0.39^**	0.47^***
ESCAPE predicts ANX				0.62^***	0.61^***				0.30^**	0.59^***
PHYS predicts ESCAPE						−1.35^*	−1.16^**	−1.56^*	−1.12^**	−1.47^**
PSYC predicts ESCAPE						1.82^**	1.97^**	2.61^**	1.68^**	2.43^**
PHYS predicts ACCEPT						—	−0.41	−2.06^**	−0.97^**	−1.87^**
PSYC predicts ACCEPT						—	0.79^**	1.88^**	1.32^**	1.80^**
Indirect effects
PSYC indirect effect to DEP									0.65^*	1.15^**
PSYC indirect effect to ANX									0.51^*	1.44^**

β = fully standardized coefficient; estimates reported are standardized coefficients and statistical significance (p) is based on z-score transformations of raw estimates.

ACCEPT, accepting responsibility coping response subscale; ANX, caregiver anxious symptoms measured by GAD-7; DEP, caregiver depressive symptoms measured by PHQ-9; ESCAPE, escape-avoidance coping response subscale; PHYS, patient physical symptom burden; PSYC, patient psychological symptom burden.

^*p < 0.05; ^**p < 0.01; ^***p < 0.001.

Relationships between coping responses and caregiver psychological distress

To evaluate the relationships between Coping Responses and Caregiver Psychological Distress, a CFA was first conducted with each of the Coping Responses factors and the Caregiver Psychological Distress factors. After the satisfactory fit of this 10 first-order factor model was confirmed, direct effects were tested.

Insufficient sample size precluded evaluation of the 16 potential pathways between Coping Responses and Caregiver Psychological Distress factors; thus, the order of input for the direct effects was determined by correlation strength. Of all the Coping Responses factors, Escape-Avoidance exhibited the strongest correlations with the Caregiver Psychological Distress factors and, therefore, was the first direct effect tested. Model 12C, which followed modifications to improve model stability and fit, suggested that FCGs' use of Escape-Avoidance coping significantly predicted the two Caregiver Psychological Distress factors: Depressive Symptoms (β = 0.63, p < 0.001) and Anxious Symptoms (β = 0.61, p < 0.001). Each of the remaining factors of Coping Responses was individually added to this direct effects model, resulting in nonsignificant findings and poorer or unchanged fit. Thus, the second hypothesis, which stated that FCGs' coping responses would predict FCGs' psychological distress, was partially supported. FCGs' use of escape-avoidance coping significantly contributed to their depressive and anxious symptoms, although the majority of their coping responses did not exert a direct effect on either their depressive or anxious symptoms.

Relationships between patient symptom management burden and coping responses

A nine-factor CFA was conducted, which included the two Patient Symptom Management Burden factors and each of the Coping Responses factors that remained in this stage of analysis. Following numerous modifications to improve fit and model stability, direct effects were tested.

The order of input of the direct effects between Patient Symptom Management Burden and Coping Responses was determined by the strength of correlations. Again, Escape-Avoidance exhibited the strongest correlations and was the first direct effect tested in this model (Table 3, Model 17B). Parameter estimates indicated that greater FCG burden due to managing Physical Symptoms predicted decreased use of Escape-Avoidance (β = −1.35, p = 0.022), whereas greater FCG burden in managing Psychological Symptoms contributed to increased use of Escape-Avoidance (β = 1.82, p = 0.009). The addition of a second direct effect Coping Response, Accepting Responsibility, produced similar results: greater FCG burden in managing Physical Symptoms predicted decreased FCG use of Accepting Responsibility (β = −2.06, p = 0.007; Model 19), whereas greater FCG burden in managing Psychological Symptoms predicted an increase in FCGs' use of an Accepting Responsibility coping response (β = 0.79, p = 0.004; Model 18B). In sum, the third hypothesis, FCGs' burden associated with managing patients' symptoms predicts FCGs' coping responses, was partially supported. FCGs reporting greater physical symptom management burden tended to use less escape-avoidance and accepting responsibility coping responses, whereas FCGs reporting greater psychological symptom management burden tended to exhibit more escape-avoidance and accepting responsibility coping.

Mediation

Escape-Avoidance was the only Coping Response that was significantly related to both a predictor variable (Psychological Symptoms) and criterion variables (Depressive Symptoms and Anxious Symptoms). Therefore, the indirect effect of Escape-Avoidance on the relationship between Psychological Symptoms and Depressive and Anxious Symptoms was tested. The parameter estimates of the indirect effects are shown in Table 3, Model 20, and are visually depicted in Figure 2. Results demonstrated that greater FCG burden in managing Psychological Symptoms significantly contributed to more Caregiver Psychological Distress, to include Depressive Symptoms (β = 0.37, p = 0.001) and Anxious Symptoms (β = 0.46, p < 0.001), and that these relationships were significantly influenced by FCGs' use of Escape-Avoidance (indirect effect on Depressive Symptoms, β = 0.65, p = 0.011, and on Anxious Symptoms, β = 0.51, p = 0.039). In sum, the fourth hypothesis (the relationship between patient symptom management burden and caregiver distress is mediated by FCGs' coping responses) was partially supported. Increased psychological symptom management burden was found to predict higher depressive and anxious symptoms among FCGs, and this relationship was partially explained by FCGs' increased use of escape-avoidance coping.

FIG. 2.

Path diagram of final mediation (Model 20, Table 3).

Discussion

The first study hypothesis (i.e., FCGs' symptom management burden predicts their psychological distress) was supported in part. Burden associated with managing patients' psychological symptoms predicted both depressive and anxious symptoms among FCGs, while burden due to managing patients' physical symptoms did not directly influence either FCGs' depressive or anxious symptoms. The lack of a predictive link between physical symptom management burden and caregiver distress was unexpected, given the body of literature suggesting that attending to patients' physical symptoms, particularly physical pain, is emotionally challenging for FCGs.^4,28–30 This may be explained, in part, by the widespread availability of medical interventions to ameliorate many physical symptoms. For example, while barriers to effective pain management are numerous,³⁰ hospice FCGs committed to reducing patients' physical pain likely have a number of highly evidence-based interventions—both pharmacological and otherwise—at their disposal.³¹ While they may be burdensome, FCGs' efforts to manage physical pain are likely to eventually be effective. Conversely, interventions to address patients' psychological symptoms (e.g., feeling sad or worrying) may be more complex, take longer to make an impact, and result in more variable degrees of success.³² Hospice clinicians should routinely assess patients' psychological symptoms and FCGs' self-efficacy as it pertains to managing those symptoms, offering support as indicated. Efforts to strengthen the evidence base for psychological interventions at the end of life, particularly brief interventions that can be delivered with limited training, should be increased.

The second study hypothesis, which stated that FCGs' coping responses would predict psychological distress, garnered nominal support. Only FCGs' use of escape-avoidance coping directly contributed to their depressive and anxious symptoms. Particularly unexpected was the lack of a significant influence of coping responses such as positive reappraisal, planful problem-solving, and seeking social support, which are fairly well-established mediators of the relationship between caregiving stressors and psychological outcomes.^7,8,33,34 To some degree, this may be due to this study's focus on coping specifically with symptom management concerns versus the broader stressors associated with care (e.g., financial disruption, role strain, and assisting with instrumental activities of daily living), or it may be linked more generally to challenges measuring complex coping processes.³⁵ Nevertheless, this finding is deserving of attention in future research.

The third study hypothesis indicated that FCGs' symptom management burden would predict the type of coping responses they would use. Results showed that higher levels of burden associated with managing patients' physical symptoms predicted less use of accepting responsibility and escape-avoidance coping. That is, FCGs experiencing high physical symptom management burden were less likely to criticize themselves or to apologize for perceived wrongdoing (possibly attributing the physical symptom management burden to the disease process itself rather than something they did), and were also less likely to rely on coping strategies such as wishful thinking or substance use. Conversely, FCGs with high degrees of burden associated with managing patients' psychological symptoms were more likely to use accepting responsibility and escape-avoidance coping. They criticized themselves (presumably finding fault with themselves) and coped by wishing things were different and/or by using food, alcohol, or medications. This suggests a number of potentially powerful opportunities to intervene. Hospice providers should provide FCGs with basic education about the psychological aspects of terminal diseases. For example, they might teach FCGs about the link between psychological and physical symptoms such as pain³⁶ or functional decline.³⁷ In addition, they should explore FCGs' beliefs about the origin of patients' psychological symptoms and invite them to reconsider understandings that place responsibility for patients' psychological well-being solely on themselves.

The fourth and final study hypothesis (i.e., coping responses mediate the relationship between FCGs' symptom management burden and their psychological distress) was partially supported. Because physical symptom management burden was not shown to predict caregiver depressive and anxious symptoms, no mediated effect in that relationship was possible. With regard to psychological symptoms, the relationship between symptom management burden and caregiver depressive and anxious symptoms was shown to be mediated by escape-avoidance coping. That escape-avoidance was the only coping response to be identified as a mediator is unsurprising, given that it was the only coping response that directly contributed to FCGs' depressive and anxious symptoms; however, this finding has important clinical implications. If FCGs were equipped with more practical and effective tools to manage patients' psychological symptoms, their reliance on escape-avoidance strategies would likely decrease, in turn reducing their experiences of depression and anxiety. FCGs should be assisted in developing active responses to patients' psychological symptoms, potentially soliciting support from hospice professionals who are trained to implement evidence-based interventions to address such symptoms. To the extent that patients' psychological symptoms are nonresponsive to clinical interventions, FCGs should be equipped with a variety of coping strategies that strengthen rather than compromise their resilience.³⁸

Study limitations

Study limitations include the lack of longitudinal data, which would have permitted an examination of family caregiving processes over time, and the geographically limited sample of participants. In addition, there was incongruence in the prescribed recall period for some of the study instruments that may have affected the accuracy of participants' responses. For example, the MSAS-SF asked which symptoms the patient had experienced in the past week, while the GAD-7 and PHQ-9 measured caregiver distress during the prior two weeks. Furthermore, caregivers who gave permission to be contacted by researchers to learn more about the study may have differed in substantive ways from those who did not. For example, overwhelmed caregivers may have withheld permission to contact, resulting in a sample of caregivers with fewer stressors or more substantive resources than the general hospice population. In addition, the mean length of hospice service for FCGs in the study was 103 days, which is substantially longer than most FCGs in the United States who receive hospice services.³⁹ While there was broad variability around the mean (SD = 185 days), FCGs in this sample had received more days of hospice care when participating in the study than most FCGs do overall. Finally, particularly when conceptualized as a complex cognitive-behavioral process, coping is notoriously difficult to measure.⁴⁰ Thus, findings that contrast sharply with the published literature on caregiver coping must be interpreted with caution and should give rise to additional inquiry in this important area of research.

Footnotes

Acknowledgments

This work was supported by the John A. Hartford Foundation (Principal Investigator: Washington; no award number). The content of this article is solely the responsibility of its authors and may not necessarily reflect the official views of the John A. Hartford Foundation.

Author Disclosure Statement

No competing financial interests exist.

References

Chentsova-Dutton

, Shucter

, Hutchin

, et al.: The psychological and physical health of hospice caregivers. Ann Clin Psychiatry, 2000; 12:19–27.

Haley

, LaMonde

, Han

, et al.: Family caregiving in hospice: Effects on psychological and health functioning among spousal caregivers of hospice patients with lung cancer or dementia. Hosp J, 2001; 15:1–18.

Parker Oliver

, Washington

, Smith

, et al.: The prevalence and risks for depression and anxiety in hospice caregivers. J Palliat Med, 2017; 20:366–371.

Parker Oliver

, Wittenberg-Lyles

, Washington

, et al.: Hospice caregivers' experiences with pain management: “I'm not a doctor, and I don't know if I helped her go faster or slower”. J Pain Symptom Manage, 2013; 46:846–858.

Johnson

, Kassner

, Houser

, Kutner

: Barriers to effective symptom management in hospice. J Pain Symptom Manage, 2005; 29:69–79.

Ratkowski

, Washington

, Craig

, Albright

: The stress of sadness: The most stressful symptoms for hospice family caregivers. Am J Hosp Palliat Care, 2015; 32:745–749.

Pearlin

, Mullan

, Semple

, Skaff

: Caregiving and the stress process: An overview of concepts and their measures. Gerontologist, 1990; 30:583–594.

Haley

, LaMonde

, Han

, et al.: Predictors of depression and life satisfaction among spousal caregivers in hospice: Application of a stress process model. J Palliat Med, 2003; 6:215–224.

Demiris

, Parker Oliver

, Wittenberg-Lyles

, et al.: A noninferiority trial of a problem-solving intervention for hospice caregivers: In person versus videophone. J Palliat Med, 2012; 15:653–660.

10.

McMillan

, Small

, Weitzner

, et al.: Impact of coping skills intervention with family caregivers of hospice patients with cancer: A randomized clinical trial. Cancer, 2006; 106:214–222.

11.

Meier

: The development, status, and future of palliative care. In: Meier

, Isaacs

, Hughes

(eds): Palliative Care: Transforming the Care of Serious Illness. San Francisco, CA: Josey-Bass, 2010, pp. 4–76.

12.

Chang

, Hwang

, Feuerman

, et al.: The Memorial Symptom Assessment Scale—Short Form (MSAS-SF). Cancer, 2000; 89:1162–1171.

13.

Lobchuk

: The memorial symptom assessment scale: Modified for use in understanding family caregivers' perceptions of cancer patients' symptom experiences. J Pain Symptom Manage, 2003; 26:644–654.

14.

Folkman

, Lazarus

: Ways of Coping Questionnaire Manual. Menlo Park, CA: Consulting Psycholgoists Press, Inc.,. 1988.

15.

Folkman

, Lazarus

: An analysis of coping in a middle-aged community sample. J Health Soc Behav, 1980; 21:219–239.

16.

Heppner

, Reeder

, Larson

: Cognitive variables associated with personal problem-solving appraisal: Implications for counseling. J Couns Psychol, 1983; 30:537–545.

17.

Kirmeyer

, Diamond

: Coping by police officers: A study of role stress and Type A and Type B behavior patterns. J Occup Behav, 1985; 6:183–195.

18.

Manne

, Sandler

: Coping and adjustment to genital herpes. J Behav Med, 1984; 7:391–410.

19.

Washington

, Rakes

: Coping responses among hospice family caregivers: A confirmatory factor analysis. J Pain Symptom Manage, 2015; 50:867–873.

20.

Spitzer

, Kroenke

, Williams

, Lowe

: A brief measure for assessing Generalized Anxiety Disorder: The GAD-7. Arch Intern Med, 2006; 166:1092–1097.

21.

Kroenke

, Spitzer

, Williams

: The PHQ-9: Validity of a brief depression severity measure. J Gen Intern Med, 2001; 16:606–613.

22.

Muthén

, Muthén

: Mplus Version 7.31 [Software]. Los Angeles, CA: Muthén & Muthén; 2015.

23.

, Bentler

: Evaluating model fit. In: Hoyle

(ed). Structural Equation Modeling: Concepts, Issues, and Applications. Thousand Oaks, CA: Sage, 1995, pp. 76–99.

24.

Goffin

: A comparison of two new indices for the assessment of fit of structural equation models. Multivariate Behav Res, 1993; 28:205–214.

25.

Steiger

: A note on multiple sample extensions of the RMSEA fit index. Struct Equ Modeling, 1998; 5:411–419.

26.

Steiger

, Shapiro

, Browne

: On the multivariate asymptotic distribution of sequential chi-square statistics. Psychometrika, 1985; 50:253–264.

27.

Allison

: Missing data techniques for structural equation modeling. J Abnorm Psychol, 2003; 112:545–557.

28.

Ferrell

: Pain observed: The experience of pain from the family caregiver's perspective. Clin Geriatr Med, 2001; 17:595–609, viii–ix.

29.

Ferrell

, Grant

, Borneman

, et al.: Family caregiving in cancer pain management. J Palliat Med, 1999; 2:185–195.

30.

Kelley

, Demiris

, Nguyen

, et al.: Informal hospice caregiver pain management concerns: A qualitative study. Palliat Med, 2013; 27:673–682.

31.

Nicholson

: Pain and its management. In: Faull

, de Caestecker

, Nicholson

, Black

(eds). Handbook of Palliative Care, 3rd ed. Hoboken, NJ: Wiley-Blackwell, 2012, pp. 109–138.

32.

Zabora

: Screening, assessment, and a problem-solving intervention for distress. In: Altilio

, Otis-Green

(eds). Oxford Textbook of Palliative Social Work. New York: Oxford University Press, 2011, pp. 169–180.

33.

Haley

, Roth

, Coleton

, et al.: Appraisal, coping, and social support as mediators of well-being in black and white family caregivers of patients with Alzheimer's disease. J Consult Clin Psychol, 1996; 64:121–129.

34.

McMillan

, Small

: Using the COPE intervention for family caregivers to improve symptoms of hospice homecare patients: A clinical trial. Oncol Nurs Forum, 2007; 34:313–321.

35.

Parker

, Endler

: Coping with coping assessment: A critical review. Eur J Pers, 1992; 6:321–344.

36.

Lee

, Wu

, Chiu

, et al.: The relationship between pain management and psychospiritual distress in patients with advanced cancer following admission to a palliative care unit. BMC Palliat Care, 2015; 14:69.

37.

Gayman

, Turner

, Cui

: Physical limitations and depressive symptoms: Exploring the nature of the association. J Gerontol B Psychol Sci Soc Sci, 2008; 63:S219–s228.

38.

Parker Oliver

, Demiris

, Washington

, et al.: Challenges and strategies for hospice caregivers: A qualitative analysis. Gerontologist, 2017; 57:648–656.

39.

Scheffey

, Kestenbaum

, Wachterman

, et al.: Clinic-based outpatient palliative care before hospice is associated with longer hospice length of service. J Pain Symptom Manage, 2014; 48:532–539.

40.

Skinner

, Edge

, Altman

, Sherwood

: Searching for the structure of coping: A review and critique of category systems for classifying ways of coping. Psychol Bull, 2003; 129:216–269.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.32 MB