Association between employment status change and depression and anxiety in allogeneic stem cell transplant caregivers

Abstract

Purpose

Allogeneic hematopoietic stem cell transplantation (Allo-HSCT) is a demanding treatment that requires caregiver support during the first 100 days post-transplant. Little is known about the association between caregivers’ employment changes and their well-being during this lengthy and challenging time when caregivers may be balancing work with caregiving responsibilities.

Methods

We used data from employed caregivers of Allo-HSCT patients who participated in two randomized clinical trials (N = 187) of an intervention for caregivers. Caregiver depression and anxiety were assessed using the Center for Epidemiologic Studies Depression and the Spielberger State-Trait Anxiety Inventory. ANCOVA was used to measure the associations between reducing work with depression and anxiety. Caregiver’s demographics and baseline employment status were controlled in the analysis along with the patient’s transplant conditioning intensity.

Results

Approximately 45% of caregivers reduced hours worked following the resumption of their caregiving duties. These caregivers, on average, had 3.49 points higher depression scores than caregivers who did not report a reduction in work (p < 0.01). Similarly, these caregivers also reported, on average, 3.56 higher anxiety scores (p < 0.10).

Conclusions

Caregivers who reduced hours worked while caregiving reported greater distress. The underlying cause of this distress deserves further investigation.

Implications for Cancer Survivors

Caregivers of patients diagnosed with cancer may experience considerable stress related to work discontinuation and caregiving. Interventions are needed to support caregivers who are trying to balance work and caregiving responsibilities.

Introduction

In 2015, approximately 39.8 million adults in the USA provided care to an adult who was disabled or ill, and among these caregivers, 2.8 million cared for a cancer patient [1-3]. The need for cancer caregivers will rise with increased survivorship associated with new treatments and targeted therapies [4-6]. Unpaid caregiving services were valued at over $450 billion/year in 2010 [2], representing more than 75% of the federal budget in 2018 for Medicare ($580 billion). Given the enormous value of caregiving services and the high level of stress associated with caregiving [4, 7], the National Cancer Institute prioritized caregiver well-being as an area for further research and interventions to alleviate their distress [8].

Caregivers of patients diagnosed with cancer commonly have elevated depression and anxiety [4, 7, 9-11]. Working-age caregivers face the additional challenge of meeting work demands while pursuing professional goals and caring for their loved one. Among the general population of cancer caregivers, 25–29% made employment adjustments to accommodate cancer caregiving [12]. In a sample of 202 employed caregivers of African American cancer survivors, 74% made an employment change and 38% reported difficulty balancing caregiving responsibilities with work [13]. Given that employment is essential for family income and/or employment-based health insurance coverage for patients and families younger than age 65 [14], it is no surprise that caregivers may desire to maintain their employment, perhaps to the detriment of their health [15]. Caregivers often come from the “sandwich generation” (those caring for spouses or partners as well as caring for their children and/or parents) [16, 17].

We report depressive and anxiety symptoms from a secondary analysis of employed caregivers of patients undergoing an allogeneic hematopoietic stem cell transplant (Allo-HSCT). Allo-HSCT is an aggressive treatment for many cancers and other conditions [18]. The decision to have an Allo-HSCT marks the beginning of intensive involvement of an informal caregiver [19] who provides supportive care during the first 100 days post-transplant. Not surprisingly, this process is extraordinarily stressful for caregivers [20], as we noted in predominately spousal caregivers of adult Allo-HSCT recipients [7, 10, 11]. The present study extends prior research by examining the relationship between employment reductions during caregiving and caregiver self-reported depression and anxiety. A reduction in hours worked may signal underlying negative situations that present opportunities for interventions to improve caregiver well-being.

Methods

Participants

Data were collected from caregivers of Allo-HSCT patients who were recruited to participate in one of two separate clinical trials that tested a behavioral stress management intervention (Psychoeducation Paced Respiration and Relaxation, PEPRR-1.0 and PEPRR-2.0) between 2008 and 2016 on caregiver report of depression and anxiety [10, 11]. Participants were recruited from the only transplant programs in the Denver metropolitan area, a university-based NCI-designated Comprehensive Cancer Center and a community-based cancer center. PEPRR-1.0 was tested in the community-based cancer center and PEPRR-2.0 was tested in caregivers at both sites. The Colorado Multi-Institutional Review Board approved both trials, registered at ClincialTrials.gov under NCT00833898 and NCT02037568.

As part of pre-transplant screening, all transplant candidates and their caregivers underwent a psychosocial interview. Caregivers were eligible if they were (1) a caregiver for an Allo-HSCT patient during the first 100 days post-transplant; (2) primarily responsible for care decisions as well as emotionally invested in the patient’s well-being; (3) able to speak and read English; and (4) 18 years of age or older. Caregivers were excluded if they had a major psychiatric disorder in the past 18 months that was unrelated to patient’s illness, or had a diagnosis of a medical condition or used medications (e.g., chronic steroid use) that might affect biomarkers collected as part of either trial. For this secondary analysis, baseline (peri-transplant) data was available from 187 employed caregivers, 98 from PEPRR-1.0 [10] and 89 from PEPRR-2.0 [11].

Assessments

Caregivers completed baseline assessments close to their patient’s Allo-HSCT (mean 2.6 ± 14.7 days prior to transplant). Data were collected on caregiver and patient characteristics, as well as caregiver employment situation, and self-reported depression and anxiety.

Demographics

Baseline demographic and clinical variables were limited to those common to both trials. These were caregiver age (continuous, years), sex (female, male), race/ethnicity (white/Caucasian/non-Hispanic, other), education (college graduate or above), annual income (< $25,000, $25,000 to $44,999, $45,000 to $64,999, and > $65,000), relationship to patient (spouse/civil partner, parent, other), patient diagnosis (leukemia, lymphoma, myelodysplastic syndrome/myeloproliferative syndrome, other), patient duration of illness (continuous, month), and patient transplant conditioning intensity (myeloablative, non-myeloablative, reduced intensity).

A reduction in employment with onset of caregiving was defined as going from full-time to part-time work, part-time work to stopping work, and full-time work to stopping work. Specific questions were: “What was your employment status BEFORE taking on caregiving responsibilities?” and “What is your CURRENT employment status as a caregiver?” Precise weekly hours worked were not collected.

Depression

The Center for Epidemiologic Studies Depression (CESD) [21] included 20 self-reported items designed to assess symptoms associated with depression, such as fatigue, restless sleep, poor appetite, and feeling lonely. Each item was scored on a 4-point Likert scale and total score ranges from 0 to 60, with higher scores indicating greater severity. A score ≥ 16 suggests depressive symptoms. Cronbach’s alpha was 0.86 for these caregivers.

State anxiety

The Spielberger State-Trait Anxiety Inventory (STAI) consists of 40 items and is a well-validated anxiety assessment [22]. For state anxiety, participants rate how they feel “right now.” Total score ranges from 20 to 80 with a higher score indicating greater anxiety. A threshold of 40 indicates significant state anxiety. Established norms for an adult population have a mean of 36. Cronbach’s alpha was 0.94 for these caregivers.

Analysis

Caregiver and patient characteristics were compared between the employed caregiver groups who reduced work with caregiving and who did not using chi-square and independent samples t-tests. Potential group differences were then evaluated for depression and anxiety scores after adjusting for demographic characteristics and patient transplant intensity using analyses of covariance (ANCOVA). Analyses were conducted using SPSS v26.0 (IBM Corporation, Armonk, NY) and SAS v9.4 (SAS Institute INC, Cary, NC) software.

Results

Caregiver and patient characteristics

Caregivers and patients from the two trials were similar except for differences between education and annual income (Appendix Tables S1 and S2). Table 1 reports sample characteristics overall and between employed caregiver groups who reduced work and those who did not. Overall, caregivers averaged 50.8 years of age and were predominantly female (76%) and white (88%). About half of the caregivers graduated from college (67%) and earned $65,000 or more annually (35%). Patient transplant condition intensity was predominantly myeloablative (57%). Nearly 72% of caregivers were employed full-time at the start of caregiving. The average CESD score was 18, indicating clinical depression. Average anxiety score was 41, suggesting mild anxiety.

Table 1.

Descriptive statistics of Employed Caregivers of Allo-HSCT Transplant Patients

	Mean (SD) or No. (%) of Sample
Characteristic	Full sample (N=187)	Work unchanged (N=102)	Reduced work (N=85)	Significance
Age (SD), y	50.8 (11.1)	50.6 (10.2)	50.9 (11.9)	p=0.85
Sex
Female	143 (76.5%)	63 (74.1%)	80 (78.4%)
Male	44 (23.5%)	22 (25.9%)	22 (21.6%)	p=0.49
Race/ethnicity
Non-Hispanic white	164 (88.2%)	76 (89.4%)	88 (86.3%)
Other	22 (11.8%)	8 ( 9.4%)	14 (13.7%)	p=0.38
Education
College graduate or above	127 (67.9%)	62 (72.9%)	65 (63.7%)	p=0.18
Annual income
<$25,000	37 (20.7%)	12 (14.1%)	25 (24.5%)
$25,000-$44,999	40 (22.4%)	22 (25.9%)	18 (17.6%)
$45,000-$64,999	39 (21.8%)	18 (21.2%)	21 (20.6%)
>$65,000	63 (35.2%)	31 (36.5%)	32 (31.4%)	p=0.23
Transplant conditioning intensity (patient)
Myeloablative	106 (56.7%)	52 (61.2%)	54 (52.9%)
Reduce intensity	53 (28.3%)	23 (27.1%)	30 (29.4%)
Non-myeloablative	18 ( 9.6%)	5 ( 5.9%)	13 (12.7%)	p=0.23
Baseline employment status
Full-time	134 (71.7%)	56 (65.9%)	78 (76.5%)
Part-time	48 (25.7%)	24 (28.2%)	24 (23.5%)	p=0.33
CESD score (SD)	17.9 (8.41)	16.0 (8.38)	19.6 (8.15)	p=0.004
State anxiety score (SD)	41.1 (11.6)	39.8 (11.5)	42.1 (11.7)	p=0.20

Notes: SD=standard deviation; CESD=Center for Epidemiologic Depression. State anxiety measured by Spielberger State Anxiety Inventory. Statistical significance determined independent t-test or Pearson's Chi-square test as appropriate between the work unchanged and reduced work samples.

In these unadjusted comparisons, only the CESD score differed statistically significantly between caregiver groups (p < 0.05), with those who reduced work reporting higher depression severity than those who did not.

Caregiver depression and anxiety

Table 2 shows that reduction in employment is significantly associated with increased depression (p < 0.01), after adjusting for participant characteristics. In this ANCOVA model, other characteristics statistically significantly associated with a higher depression score were female (4.32, p < 0.01) and absence of a college degree (2.85, p < 0.10); lower scores were associated with age (−0.09, p < 0.10) and non-myeloablative (−5.40, p < 0.05).

Table 2.

Adjusted Relationships of Reduced Employment in Caregivers with Depression and Anxiety Adjustment, N=187

Characteristic	CESD Score	Spielberger State Anxiety Score
Reduced employment
No	Reference	Reference
Yes	3.49 (1.25)***	3.35 (1.82)*
Age	−0.09 (0.06)*	0.01 (0.08)
Sex
Male	Reference	Reference
Female	4.32 (1.48)***	3.56 (2.16)
Race/ethnicity
Other	Reference	Reference
Non-Hispanic white	2.77 (1.97)	5.82 (2.80)**
Education
College graduate	Reference	Reference
No college degree	2.85 (1.53)*	0.66 (2.18)
Annual income
<$25,000	Reference	Reference
$25,000-$44,999	0.97 (2.00)	5.18 (2.91)*
$45,000-$64,999	0.14 (2.13)	1.22 (3.09)
>$65,000	0.30 (2.16)	3.19 (3.06)
Transplant conditioning intensity (patient)
Myeloablative	Reference	Reference
Reduce intensity	−3.21 (1.38)	−2.64 (2.00)
Non-myeloablative	−5.40 (2.11)**	−2.64 (3.08)
Baseline employment status
Full-time	Reference	Reference
Part-time	0.61 (1.61)	−0.81 (2.29)

Notes: CESD=Center for Epidemiologic Depression. Standard errors shown in parentheses. Statistical significance reported at

^*p< 0.10

^**p<0.05

^***p<0.01.

There was a trend for reduction in employment associated with increased anxiety scores (p < 0.10) from Spielberger State-Trait Anxiety Score, after adjusting for participant characteristics, In this ANCOVA model, other characteristics statistically significantly associated with higher anxiety scores were non-Hispanic White (5.82, p < 0.05) and a trend for earning between $25,000 and $44,999 relative to caregivers earning < $25,000 (p < 0.10).

Discussion

Caregiving for a cancer patient may come with serious consequences for caregivers [8, 23]. To accommodate caregiving, caregivers may have to alter their employment status. In a study of 5000 patients, 25–29% of their caregivers took extended leaves from work [12]. In our study, 45% of the employed caregivers reported a reduction in employment, perhaps reflecting the intensive nature of Allo-HSCT caregiving [24, 25]. Caregivers who reduced work had higher levels of depression and anxiety, often exceeding thresholds for clinical concern.

In most households, employment is essential for family income and/or employment-based health insurance coverage particularly for patient groups under age 65 years [14]. It comes as no surprise that caregiving may interfere with work performance, stress, and health of the caregiver [15]. At the household level, health shocks (e.g., any significant change in health) [26] can influence patients’ retirement decisions and whether caregivers remain in the work force to guarantee income and health insurance [14].

The association between caregiving stress and health [27] led former US Surgeon General to recommend a “Rx for Caregivers” that included attending to depression and anxiety, identifying sources of support, reducing stress, while taking care of one’s physical health [28]. There are an increasing number of supportive interventions directed toward caregivers of patients diagnosed with cancer offering innovative methods to improve quality of life [29], although caregivers’ and Allo-HSCT patients’ distress may be elevated over an extended period [30] increasing the need for efficacious interventions [31, 32] that convincingly reduce caregiver distress [10, 11]. Interventions that provide care and support focused on the needs of employed caregivers are required and should be tailored to their caregiving responsibilities such as developing web-based interventions [33, 34].

Implications

Caregivers often must make tough employment decisions when caring for a loved one. Reducing work can range from early retirement to deferring career advancement in lieu of supporting their loved one through the cancer experience. Employers need to be fully cognizant of the distress experienced by employees serving in an informal caregiving role. We noted that a reduction in work was associated with greater symptoms of depression and anxiety among employed caregivers. Furthermore, the overall mean depression score (18) and mean score for caregivers who reduced work (20) exceeded the generally accepted cut-off of 16 on the CESD for individuals at risk for clinical depression [21]. Developing scalable interventions for employed caregivers will be the target of future studies.

Limitations

This secondary analysis has three main limitations. First, we did not record the exact number of hours worked before and after transplant. Second, our findings do not generalize to other cancer caregivers or caregivers in general. In addition, the caregivers of Allo-HSCT patients in the present study were mostly women (76%), and their experiences may not generalize to men. As with most secondary analyses, we report associations and cannot make causal inferences. The study lacks information on patients’ disease severity and complications. It is conceivable that as patients’ health deteriorated, caregivers had to stop working to provide more intensive care. Stress and anxiety may stem from concern for patients as well as changes in work. Other conditions in the household may also contribute to the increased depression and anxiety we observed.

Conclusions

There is a significant gap in the literature regarding the challenges employed caregivers experience, highlighting the need to support members of this important population who are distressed by both work and caregiving responsibilities. Distress was significantly greater in caregivers who reduced work compared to those who did not. There is a need to further study and understand the underlying mechanism(s) of depression and anxiety associated with caregiver work reduction, which may be related to loss of job opportunities, income, and/or benefits, loss of workplace companionship, increased burden created when balancing work and caregiving demands, or any number of other possibilities that influence day-to-day functioning. An improved understanding of these stressors will lead to interventions that mitigate the adverse impact of caregiving on those employed. New evidence could lead to policy and workforce changes that support caregivers more comprehensively than presently available. We view this study as a step toward highlighting the unmet needs of caregivers who provide invaluable services to their loved ones and society.

Data availability

All primary data reported is available for review on request from the corresponding author following removal of all identifying information.