Abstract
Background
Family caregivers are at a high risk for low quality of life due to caregiving-related stress. Caregivers' stress is commonly assessed using self-reported measures, which reflect relatively subjective and long-term stress related to caregiving, but objective biological markers of stress are rarely used for caregivers. The purposes of this study were (1) to determine whether caregiver characteristics were associated with stress assessed using a stress biomarker (serum cortisol) and a self-reported caregiving distress measure (Caregiver Burden Inventory) and (2) to determine the predictability of both stress measures for quality of life in caregivers of patients with heart failure.
Methods
Taiwanese family caregivers (N = 113; mean age, 54.5 years; 70.8% female) of patients with heart failure completed surveys including caregiving distress and quality of life measured by the Caregiver Burden Inventory and the Short Form-36 (physical and psychological well-being subscales), respectively, and provided blood samples for serum cortisol. Independent t tests, correlation, and hierarchical regression were conducted.
Results
Single caregivers had higher serum cortisol levels than married caregivers (P = .002). Men had significantly higher serum cortisol levels than women (P = .010), but men reported lower caregiving distress than women (P = .049). Both serum cortisol (β = −0.32, P = .012) and caregiving distress (β = −0.29, P = .018) were significant predictors of quality of life in the physical well-being scale while controlling for caregivers' characteristics and depressive symptoms. Serum cortisol (β = −0.28, P = .026) and caregiving distress (β = −0.25, P = .027) also predicted quality of life in the psychological well-being scale.
Conclusions
Serum cortisol and self-reported caregiving distress have similar predictability for quality of life in family caregivers of patients with heart failure. Reducing stress and caregiving distress is critical to improving quality of life in this population.
KEY WORDS: caregiving distress, heart failure, quality of life, serum cortisol level
Heart failure (HF) is characterized by a gradual decline in health interspersed with episodes of acute deterioration and, often, sudden unexpected death.1 Family caregivers' support and contribution to HF management substantially improve patients' health outcomes.2,3 On the other hand, because of stress or burden related to caregiving responsibilities, caregivers are known to be a vulnerable population at a high risk of poor physical and psychological health (eg, depressive symptoms, cardiovascular disease development, and mortality)4,5 and poor quality of life.6
Caregivers generally have a worse quality of life than noncaregivers.7 Caregivers who care for patients with HF commonly report poor quality of life,8 and their quality of life is even lower than that of caregivers who care for patients with other chronic diseases.9 In 2 patient-caregiver dyadic studies in HF, caregivers experienced poor quality of life that was similar in severity to that of patients with HF.6,10 According to a meta-analysis of 15 HF dyadic studies, caregivers' perceived strain and psychological distress assessed by self-reported measures were significantly associated with poor quality of life in patients.11 Such evidence indicates that it is important to identify caregivers with a greater risk of stress and low quality of life.
In managing HF, caregiver responsibilities include ensuring that patients take medication as prescribed, monitoring HF symptoms, and managing medical and diet regimens. Caregivers also commonly deal with frequent patient hospitalizations, complex treatment devices, and the financial burden of HF.12,13 Stress related to caregiving responsibilities is commonly called caregiving distress or burden. Most research focused on measuring caregivers' stress includes various self-reported instruments (eg, Caregiver Burden Inventory [CBI]).14–16 These subjective measures focus on long-term stress (eg, 1 week, 2 weeks, or 1 month) rather than capturing the everyday stress experience.
Biological markers have rarely been used to assess stress in caregivers and have not been used at all in caregivers of patients with HF. Exposure to a stressor stimulates the hypothalamic-pituitary-adrenal axis, increasing the secretion of stress-related glucocorticoid hormones, including cortisol. The hypothalamic-pituitary-adrenal axis is one of the most important neuroendocrine stress systems. Dysregulation in the hypothalamic-pituitary-adrenal axis is implicated in the pathophysiology of stress-related disorders such as depression and anxiety. Cortisol is considered an objective biological marker that quantifies the amount of stress an individual perceives themselves to be experiencing.17 Serum cortisol levels reveal acute changes at a single point in time and may be influenced by various short-term factors, including recent stressors or activities. To assess long-term or chronic stress accurately, a more comprehensive approach (eg, a series of cortisol measurements over time) is required. Overall long-term systemic cortisol exposure is difficult to evaluate due to circadian variations and its protein-binding capacity.18 Serum cortisol levels can fluctuate throughout the day and generally increase during the early morning (ie, highest at about 8 am) and decrease slightly in the evening and during the early phase of sleep.19 Cortisol, assessed by either serum or salivary samples, has been used to measure physiologic stress for caregivers who care for patients with other chronic diseases such as dementia and cancer,20,21 but not in caregivers of patients with HF.
Despite the stressful nature of caregiving responsibility, there is a lack of research to assess caregivers' stress using biomarkers, including cortisol, to predict the quality of life in caregivers. Researchers suggest that evaluating both cortisol levels and subjective stress related to caregiving could help identify groups with poor responses to stress and elevated cortisol levels.22 However, it is unknown whether either long-term subjective stress related to caregiving (ie, caregiving burden in self-report), short-term objective biological markers of stress in caregivers, or both can predict quality of life in family caregivers of patients with HF. In this light, the study's specific aims were (1) to examine whether caregiver characteristics were associated with stress assessed using a stress biomarker (ie, serum cortisol) and a self-reported caregiving distress measure (ie, CBI) and (2) to determine whether both measures, the stress biomarker and self-reported caregiving distress, predict the quality of life (ie, physical and psychological well-being) in family caregivers of patients with HF.
Methods
Study Design and Sample
We used a cross-sectional design to investigate family (ie, informal, unpaid) caregivers of patients with HF. Eligible caregivers were the primary adult caregivers (older than 21 years) who cared for a patient (care recipient) with a confirmed HF diagnosis verified by cardiologists. We excluded caregivers caring for a family care recipient with a history of cerebral vascular accidents, recent myocardial infarction (within the previous 6 months), or a major comorbid terminal illness, or those receiving advanced treatment such as hemodialysis or concurrent cancer treatment. We also excluded caregivers caring for a patient/care recipient with cognitive impairment (eg, dementia) or residing in institutional settings. Caregivers with major health problems, such as cancer, recent myocardial infarction, or terminal disease, were excluded from the study. We recruited caregivers who visited outpatient HF clinics with their care recipients at an academic medical center in southern Taiwan.
Data Collection Procedure
After approval to conduct the study from the institutional review boards, eligible participants were approached and recruited at the academic medical center outpatient clinic. Participants who agreed to participate in the study signed a written informed consent form after receiving comprehensive information about the purposes and procedures of the study. We scheduled a date for participants to visit the research site for data collection. When eligible participants arrived on the designated morning, they completed a standard questionnaire. A blood sample for serum cortisol was drawn by trained staff. The entire procedure lasted 20 minutes on average. Demographic and clinical data of care recipients were collected via chart review.
Measures
Serum Cortisol
Serum cortisol level is known as an objective biomarker for stress.23 We obtained 3 to 5 mL of blood using venipuncture. We used specific time-point cortisol by collecting blood samples between 9 am and noon on the scheduled date. Serum cortisol levels were then assessed using the chemiluminescence immunoassay technique and the enzyme-linked immunosorbent assay (ADVIA Centaur Cortisol assay kit; Siemens, Camberley, United Kingdom) in an academic medical center laboratory. The intra-assay and interassay coefficients of variation for serum cortisol results were less than 10%, and the sensitivity limit was 0.20 μg/dL. The reference range of blood cortisol in the morning is 4.82 to 19.5 μg/dL. A high level of serum cortisol indicates a high level of stress. Cortisol levels can be assessed in urine, saliva, or blood samples. Compared with salivary and urinary cortisol, serum cortisol is considered more reliable for measuring the secretion of stress-related glucocorticoid hormones. This is because renal function or medication less affects serum cortisol levels, providing a more accurate reflection of the body's stress response.22,24 The current study's decision to use serum cortisol was motivated by 2 main reasons. First, it allows us to minimize the time gap between the self-report survey and the stress biomarker. Second, the convenience of data collection played a significant role, because obtaining urine samples (ie, 24-hour urine) or saliva samples for cortisol tests (ie, at least 2 days' samples) can be relatively burdensome and time consuming for study participants. By opting for serum cortisol, we aimed to streamline the data collection process and enhance participant compliance.
Subjective Stress Related to Caregiving
In this study, we operationalized subjective stress related to caregiving as self-reported caregiving distress and used the CBI to measure it. The CBI, developed by Novak and Guest,25 has been widely used for assessing caregiving distress in caregivers of patients with HF.26 We assessed caregiving distress using the Chinese version of the CBI validated for Taiwanese caregivers.25,27 The CBI consists of 24 items and is rated on a 5-point Likert scale from 0 (strongly disagree) to 4 (strongly agree) based on distress over the past month. Although the CBI has 5 subscales, we used the total score in this study. The total score is the sum of the 24 items, and it can range from 0 to 96, with higher scores indicating higher levels of perceived stress related to caregiving. High correlations supported the 2-week test-retest reliability of the Chinese version of the CBI for 5 dimensions (0.88–0.93), and adequate internal consistency was demonstrated with Cronbach αs ranging from 0.79 to 0.93.28 The construct validity of the CBI was supported by testing known hypotheses and with factor analysis.28 Internal consistency reliability was supported by Cronbach αs of 0.94 for the 24 items in the current study.
Quality of Life
Health-related quality of life was assessed using the Short Form-36 Health Survey Questionnaire Taiwan version, which demonstrates good reliability (Cronbach α = 0.86) and validity in a general Taiwanese population.29 The Short Form-36 generated 2 summary scores of quality of life, physical and psychological well-being, that were calculated from 36 items by using norm-based Taiwan-specific Short Form-36 algorithms.28,30 The raw scores of the 2 subscales were transformed into a 0-to-100 scale, and a higher score indicates better health-related quality of life.
Sample Characteristics and Potentially Confounding Factors
We collected self-reported demographic information of caregivers (eg, age, education level, gender, marital status, living arrangement, relationship with the care recipient, and having a chronic illness) to describe sample characteristics. Because we excluded caregivers with major health problems (eg, cancer, recent myocardial infarction, or terminal disease), the chronic illnesses of caregivers in the questionnaire included hypertension, diabetes, hyperlipidemia, and heart disease. We also collected care recipients' sociodemographic data (eg, age, gender, education, marital status, employment), length of HF diagnosis, left ventricular ejection fraction, and New York Heart Association class to describe care recipients' characteristics.
We collected caregivers' depressive symptoms as a potential confounding factor. Depressive symptoms are prevalent and a well-known predictor of quality of life in caregivers.31,32 Depressive symptoms were assessed using the Patient Health Questionnaire-9 (PHQ-9). The PHQ-9 is a depression instrument that has items that each correspond to 9 criteria for depression in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition.33 The PHQ-9 assesses an individual's emotional condition over the last 2 weeks, including anhedonia, depressed mood, insomnia or hypersomnia, fatigue or loss of energy, appetite disturbances, feelings of guilt or worthlessness, diminished ability to think or concentrate, psychomotor agitation or retardation, and suicidal thoughts. All items are rated from 0 (not at all) to 3 (nearly every day). Summing the scores of all 9 items gives a total score that ranges from 0 to 27, with higher scores indicating worse depression. The validity of the PHQ-9 for detecting depression has been assessed and established over the last few decades in large population samples.34 The Chinese version of the PHQ-9 has been well validated in Chinese populations.34–36 In the current study, internal consistency reliability was supported by Cronbach αs of 0.91.
Data Analyses
Descriptive statistics were conducted to describe the characteristics of caregivers and care recipients in this study. For the first specific aim, we used independent sample t tests or Pearson correlations to examine associations of the characteristics of caregivers with objective and subjective stress measures (ie, serum cortisol and the CBI), as appropriate. For the second specific aim, after we reviewed the assumptions for multiple regression (ie, multicollinearity, normal distribution), 2 hierarchical multiple linear regressions were conducted to determine predictors of quality of life using the physical and psychological well-being scores. In the hierarchical regression model, covariates were as follows: First, we added caregivers' sociodemographic characteristics (ie, age, gender, education level, marital status, and relationship with the care recipient) and depressive symptoms in the first block. We added serum cortisol and caregiving distress in the second block. Before performing regression analysis, we examined correlations among primary independent variables and variance inflation factors to identify multicollinearity problems. We also examined the assumption of linearity, homoscedasticity, and normality of residuals and found that those assumptions were met. During the examination of multicollinearity, we found no significant multicollinearity issues, as indicated by the variance inflation factor values ranging from 1.2 to 4.5 (all Ps > .05) for predicting physical and psychological well-being. The correlation values among independent variables were less than 0.70, further confirming the absence of multicollinearity. In addition, the P values obtained from the Kolmogorov-Smirnov normal distribution test were all greater than .05, suggesting that the data followed a normal distribution.
We estimated the sample size using G*Power before conducting the study. A sample size of 54 participants was necessary for running a multiple regression analysis with 8 predictors, considering an effect size of 0.5,37 a power of 0.8, and an α of 0.05. In this study, we successfully recruited 113 participants, which ensures having adequate power for detecting effects. All data analyses were conducted using SPSS 27 (IBM SPSS Inc, Armonk, New York).
Results
Sample Characteristics
Of the 125 eligible caregivers we approached, 9 caregivers declined to participate in the study, and 3 caregivers could not provide serum cortisol blood samples because of scheduling conflicts; thus, a total of 113 caregivers agreed to participate and provided the complete data for this study. There were no participants who dropped out during data collection. Table 1 presents the characteristics of caregivers and care recipients. The mean age of caregivers was 54.5 years. Most caregivers were female (71%), living with the care recipients (74%), and married/cohabitating (78%). Most caregivers (89%) took care of their spouse or ex-spouse. One-third of caregivers reported they have at least 1 chronic illness, including hypertension, diabetes, and hyperlipidemia most commonly. The mean (SD) score on the PHQ-9 for caregivers was 8.38 (6.89).
TABLE 1.
Sample Characteristics (N = 113)
| Characteristics | Mean (SD) or n (%) |
|---|---|
| Caregivers | |
| Age, y | 54.48 (13.25) |
| Education, y | 11.25 (3.70) |
| Gender | |
| Male | 33 (29.2%) |
| Female | 80 (70.8%) |
| Marital status | |
| Single/divorced/widowed | 25 (22.1%) |
| Married/cohabitating | 88 (77.9%) |
| Living with care recipient | |
| Yes | 83 (73.5%) |
| No | 30 (26.5%) |
| Relationship with care recipient | |
| Spouse/ex-spouse | 100 (88.5%) |
| son/daughter (in-law) | 8 (7.1%) |
| Others | 5 (4.4%) |
| Having chronic illness | |
| No | 75 (66.4%) |
| Yes | 38 (33.6%) |
| Depressive symptoms, PHQ-9 score | 8.38 (6.89) |
| Care recipients | |
| Age, y | 69.20 (13.15) |
| Gender | |
| Male | 60 (53.1%) |
| Female | 53 (46.9%) |
| Marital status | |
| Single/divorced/widowed | 30 (26.5%) |
| Married/cohabitating | 83 (73.5%) |
| Employment | |
| No | 100 (88.5%) |
| Yes | 13 (11.5%) |
| Length of heart failure diagnosis, mo | 58.15 (56.21) |
| Left ventricular ejection fraction | |
| <40% | 51 (45.1%) |
| ≥40% | 62 (54.9%) |
| New York Heart Association class | |
| I/II | 46 (40.7%) |
| III/IV | 67 (59.3%) |
Abbreviation: PHQ-9, Patient Health Questionnaire-9.
Half of the care recipients were male (53%) and in New York Heart Association class III/IV (59%) and had a reduced ejection fraction of less than 40% (n = 51, 45.1%). Most had a partner (74%) and were unemployed (89%). The mean (SD) age of care recipients was 69.2 (13.2) years, and the mean (SD) length since diagnosis was 58 (56.4) months.
Associations Between Stress and Characteristics of Caregivers
The mean (SD) serum cortisol level was 9.59 (4.57) μg/dL (median, 9.0 μg/dL). The 25th and 75th quartiles of serum cortisol levels were 5.91 and 11.42 μg/dL, respectively. The mean (SD) CBI score was 34.66 (15.23). Tables 2 and Table 3 present the results of bivariate associations between caregivers' characteristics and stress measures (ie, serum cortisol and caregiving distress) and correlations among variables. Education in years was positively correlated with serum cortisol (r = 0.23, P = .015) but not significantly correlated with caregiving distress. Male caregivers had significantly higher serum cortisol levels than female caregivers (10.69 vs 8.47 μg/dL, P = .010). In contrast, female caregivers reported higher caregiving distress than male caregivers on the self-reported CBI (1.60 vs 1.30, P = .049). Single caregivers had more elevated serum cortisol levels than married/cohabitating caregivers (11.44 vs 8.51 μg/dL, P = .002). However, the 2 groups had no significant difference in caregiving distress. Depressive symptoms (ie, PHQ-9 scores) were significantly correlated with caregiving distress (r = 0.32, P < .001), but not correlated with serum cortisol. Caregivers' age and chronic illnesses as well as their relationship and living status with the care recipient were associated with neither serum cortisol nor caregiving distress. We also found no significant correlation between serum cortisol and self-reported caregiving distress (r = −0.13, P = .162).
TABLE 2.
Association of Caregiver Characteristics With Serum Cortisol and Caregiving Burden (N = 113)
| Characteristics of Caregivers | Serum Cortisol | Caregiving Distress | ||
|---|---|---|---|---|
| Mean (SD) | t Test (P) | Mean (SD) | t Test (P) | |
| Gender | 2.61 (.010) | 1.99 (.049) | ||
| Male | 10.69 (3.94) | 1.30 (0.83) | ||
| Female | 8.47 (4.29) | 1.60 (0.64) | ||
| Marital status | 3.13 (.002) | −1.60 (.113) | ||
| Single | 11.44 (4.63) | 1.71 (0.75) | ||
| Married/cohabitating | 8.51 (3.98) | 1.45 (0.73) | ||
| Relationship with care recipient | 2.20 (.116) | 1.18 (.311) | ||
| Wife/husband | 8.28 (4.05) | 1.38 (0.65) | ||
| Son/daughter (in-law) | 10.01 (4.42) | 1.59 (0.81) | ||
| Others | 9.48 (3.08) | 1.65 (0.70) | ||
| Living with patient | 1.17 (.243) | 0.29 (.773) | ||
| No | 9.93 (4.21) | 1.48 (0.70) | ||
| Yes | 8.84 (4.31) | 1.52 (0.76) | ||
| Having chronic illness | −0.37 (.712) | −0.13 (.897) | ||
| No | 9.06 (4.66) | 1.51 (0.76) | ||
| Yes | 9.35 (3.55) | 1.49 (0.71) | ||
TABLE 3.
Correlations Among Caregiver Characteristics, Serum Cortisol, and Caregiving Burden
| Caregiver Characteristics | Age | Education | Depressive Symptoms | Serum Cortisol | Caregiving Distress |
|---|---|---|---|---|---|
| Age | — | ||||
| Education | −0.47a | — | |||
| Depressive symptoms (PHQ-9) | −0.12 | −0.10 | — | ||
| Serum cortisol | 0.16 | 0.23b | 0.07 | — | |
| Caregiving distress | −0.07 | −0.14 | 0.33a | −0.13 | — |
Abbreviation: PHQ-9, Patient Health Questionnaire-9.
aP < .001.
bP < .05.
Predictors of the Quality of Life
The mean (SD) physical well-being score was 49.7 (8.1; range, 37.9–71.2), indicating lower levels of physical well-being. The mean (SD) psychological well-being score was 50.3 (10.5; range, 29.7–72.7), indicating lower levels of psychological well-being. Table 4 presents the results of hierarchical multiple regression analyses predicting physical and psychological well-being.
TABLE 4.
Predictors of Caregivers' Physical and Psychological Well-being (N = 113)
| Variable | Physical Well-being | Psychological Well-being | ||||||
|---|---|---|---|---|---|---|---|---|
| B | SE | β | P | B | SE | β | P | |
| Model 1 | ||||||||
| Age | −0.06 | 0.11 | −0.09 | .59 | 0.14 | 0.13 | 0.18 | .274 |
| Gendera | 1.94 | 2.41 | 0.10 | .42 | −0.86 | 2.93 | −0.03 | .771 |
| Education | −0.65 | 0.33 | −0.34 | .04 | 0.26 | 0.40 | −0.10 | .527 |
| Marital statusb | −4.60 | 2.51 | −0.24 | .07 | −0.58 | 3.06 | −0.03 | .850 |
| Relationship with care recipientc | −1.50 | 1.76 | 0.23 | .40 | −0.64 | 2.14 | −0.04 | .765 |
| Depressive symptoms | −4.24 | 1.26 | −0.40 | .001 | −6.07 | 1.54 | −0.47 | <.001 |
| Model 1 summary | R2 = 0.264, adj R2 = 0.182, F = 3.337, P = .007 | R2 = 0.284, adj R2 = 0.208, F = 3.764, P = .003 | ||||||
| Model 2 | ||||||||
| Age | −0.09 | 0.10 | −0.13 | .39 | 0.11 | 0.12 | 0.14 | .360 |
| Gendera | 1.72 | 2.42 | 0.09 | .48 | −2.27 | 2.99 | −0.09 | .451 |
| Education | −0.68 | 0.32 | −0.32 | .04 | 0.31 | 0.39 | 0.11 | .440 |
| Marital statusb | −6.21 | 2.46 | −0.33 | .01 | −2.38 | 3.04 | −0.10 | .437 |
| Relationship with care recipientc | −0.98 | 1.70 | −0.07 | .57 | −0.53 | 2.10 | −0.03 | .801 |
| Depressive symptoms | −3.47 | 1.23 | −0.47 | .006 | −5.42 | 1.52 | −0.42 | .001 |
| Serum cortisol | −0.38 | 0.22 | −0.32 | .012 | −0.62 | 0.27 | −0.28 | .026 |
| Caregiving distress | −3.36 | 1.38 | −0.29 | .018 | −1.98 | 1.71 | −0.25 | .027 |
| Model 2 summary | R2 = 0.379, adj R2 = 0.269, F = 1.68, P = .041 | R2 = 0.420, adj R2 = 0.363, F = 3.806, P = .001 | ||||||
Abbreviations: adj, adjusted; B, unstandardized coefficient; β, standardized coefficient.
aReference group: male.
bReference group: single.
cReference group: wife/husband.
Physical Well-being
Control variables (ie, age, gender, education, marital status, relationship with the care recipient, and depressive symptoms) in the first block explained 26.4% of the variance in physical well-being. After entering the 2 stress measures in the second block, a total of 37.9% of the variance in caregivers' physical well-being was explained. The 2 stress measures, serum cortisol (β = −0.32, P < .05) and caregiving distress (β = −0.29, P < .05), were significant predictors of physical well-being and explained an additional 11.5% of the variance.
Psychological Well-being
The control variables in the first block explained 28.4% of the variance in caregivers' psychological well-being. In the second block, serum cortisol (β = −0.28, P < .05) and caregiving distress (β = −0.25, P < .05) were also independent predictors of psychological well-being and explained an additional 13.6% of the variance in psychological well-being. The full models explained a total of 42.0% variance in psychological well-being.
Discussion
The most compelling finding in this study was that both serum cortisol and self-reported caregiving distress were independent predictors of quality of life in the physical and psychological well-being domains, and the 2 measures had similar predictability for the 2 outcomes. The findings in this study were robust because we controlled the potential covariates of caregivers' characteristics as well as their depressive symptoms, which are common and well-known factors associated with quality of life in caregivers of patients with HF.38,39 Thus, our findings suggest that managing and relieving caregiving stress is vital to improving the quality of life in the physical and psychological well-being of family caregivers of patients with HF. In future research, it would be valuable to examine the effectiveness of interventions targeting cortisol levels and subjective reports on caregiving stress to improve the quality of life among caregivers of patients with HF.
The strength of this study is the use of both objective and subjective measures of caregiver stress to predict caregivers' quality of life. In most caregiver studies, biomarkers have rarely been used to assess caregivers' stress. Our findings may support the use of serum cortisol by future researchers as a reliable objective measure to assess caregiver stress in caregiving science. A noticeable finding was that the 2 measures (ie, serum cortisol and self-reported caregiving distress) were not correlated. This result could be explained by the fact that these 2 measures capture somewhat distinct aspects of caregiver stress, as we expected. Serum cortisol, which was assessed only once in this study, likely represents an immediate physical responses and is relatively influenced by various short-term factors, including short-term stress. In contrast, self-reported caregiving distress measured by the CBI reflects a broader evaluation of psychological and social factors associated with longer-term caregiving stress.
Interestingly, these 2 stress measures were associated with caregivers' gender, but in opposite directions, because male caregivers had higher serum cortisol levels and lower self-reported caregiving distress than female caregivers. Our findings are similar to the results of previous studies.40,41 According to a meta-analysis of 65 studies, gender differences in salivary cortisol were significant, and men's salivary cortisol responses were 1.5 times higher than those of age-matched women when facing stress.41 In our study, we added evidence that serum cortisol levels differed by gender.
In our study, female caregivers reported higher caregiving distress than male caregivers. The caregiver literature robustly supports the evidence that female caregivers are more burdened and distressed than male caregivers in self-report. In a study with a large sample (N = 1611), Swinkels et al40 reported that female caregivers experienced a greater burden than male caregivers. They also reported that female caregivers experienced more secondary stressors than male caregivers, whereas caregiving intensity was a primary contributor to a greater burden among male caregivers. Female caregivers in Taiwan also reported experiencing secondary stressors such as limited support from other family members of care recipients, family obligations, cultural expectations, social norms, worries about the future, unemployment, financial problems, and care resource shortage.42–44 Female caregivers may have different perceptions of stress and caregiving distress that affect quality of life differently than male caregivers. Because most participants in our study were female, further investigation is needed to recruit an equal proportion of male and female caregivers and determine whether the gender of caregivers moderates the association between caregiver stress and their quality of life.
Another noticeable finding was that serum cortisol was associated with marital status, in that single caregivers had higher serum cortisol levels than married/cohabitating caregivers. Single caregivers needing to care for ill family members may face more stress than married/cohabitating caregivers, which is a growing issue in Taiwan.43 In today's Taiwanese society, a single person in the family often cares for their ill siblings or parents.45,46 It is also possible in Taiwanese culture that when a person is responsible for caring for a sick family member, he/she might decide to delay their marriage. The fact that family caregivers may not have adequate knowledge, skills, training, time, income, assistance, and resources to meet the needs of the care recipients, particularly when single, can be a major cause of caregiving stress.46,47 Moreover, in Taiwanese culture, most married individuals live with or near extended families where they can share their stresses and responsibilities with other family members and receive more family support than single individuals. Therefore, healthcare providers must screen single caregivers' support systems and offer resources to help them cope with stressful situations.
This study has several limitations. Because this study is a cross-sectional study, interpreting the causal relationship between stress measures and quality of life should be done cautiously. In addition, the one-time measure of cortisol during the morning may not reflect the overall volume of cortisol related to the waking period. Nevertheless, the one-time collection of serum cortisol in this study had strong predictability for quality of life in our study. Because participants in this study were recruited from southern Taiwan, generalizability may be limited to caregivers in Western or other Asian countries. In addition, we did not collect caregivers' clinical conditions (eg, mediations, sleep habits, prescribed mediation) that may influence caregivers' cortisol levels, although we excluded caregivers with major health problems. Further investigation into controlling caregivers' medical conditions that may influence serum cortisol levels will support our findings. Finally, we solely relied on caregiver-related variables, omitting care recipients' variables, to investigate the proposed specific aims because this study focused exclusively on caregiver-related variables. While acknowledging the significance of the interaction between caregivers and care recipients, our study design limited the scope to caregiver-centric aspects.32 Future research should consider exploring the reciprocal nature of the caregiver and care recipient relationship and examining the potential influence of care recipient variables, particularly those related to stress, on caregivers' quality of life.
Conclusion
The perceived stress and caregiving distress were associated with the caregivers' characteristics, and the association direction was opposite by the gender of caregivers. The biomarkers of stress and self-reported caregiving distress have similar predictability for quality of life in family caregivers of patients with HF. Thus, reducing stress and caregiving distress is necessary to improve the quality of life in this population.
What’s New and Important
Serum cortisol is a short-term, objective measure and predictor of stress overall. In contrast, self-reported caregiving distress is a subjective measure assessing long-term perceived stress related to caregiving responsibility.
The strength of this study is the use of objective and subjective measures of caregiver stress to predict caregivers' health-related quality of life.
The biomarkers of stress and self-reported caregiving distress have similar predictability of quality of life in family caregivers of patients with HF.
Footnotes
This study was supported by the Ministry of Science and Technology in Taiwan (104-2314-B-255-006-MY3) and Chang Gung Medication Foundation (CMRPG8D1531).
The authors have no funding or conflicts of interest to disclose.
Contributor Information
Ya-Ching Liu, Email: ching5345@cgmh.org.tw.
Man-Chun Chou, Email: anny5626@cgmh.org.tw.
Ming-Chu Chiang, Email: e2988386@cgmh.org.tw.
Chih-Ling Hang, Email: samuelhang2000@yahoo.com.
Shyh-Ming Chen, Email: Syming99@gmail.com.
Misook L. Chung, Email: misook.chung@uky.edu.
Chin-Yen Lin, Email: cy.lin@uky.edu.
REFERENCES
- 1.Ponikowski P Anker SD AlHabib KF, et al. Heart failure: preventing disease and death worldwide. ESC Heart Fail. 2014;1(1):4–25. [DOI] [PubMed] [Google Scholar]
- 2.Buck HG Harkness K Wion R, et al. Caregivers' contributions to heart failure self-care: a systematic review. Eur J Cardiovasc Nurs. 2015;14(1):79–89. [DOI] [PubMed] [Google Scholar]
- 3.Su Y Cheng M Zhu C, et al. The influence of informal caregivers' preparedness on psychological symptoms and quality of life among patients with heart failure and insufficient self-care. J Cardiovasc Nurs. 2023;38(3):224–236. [DOI] [PubMed] [Google Scholar]
- 4.Ahn S, Esquivel JH, Davis EM, Logan JG, Chung ML. Cardiovascular disease incidence and risk in family caregivers of adults with chronic conditions: a systematic review. J Cardiovasc Nurs. 2022;37(3):E47–E60. [DOI] [PubMed] [Google Scholar]
- 5.Dong X, He D, Zhang Y, Zhao Q, Zhang X, Fan X. Dyadic associations between burden and depressive symptoms among patients with heart failure and their caregivers: the mediating role of perceived stress. J Cardiovasc Nurs. 2023. doi: 10.1097/jcn.0000000000000974. [DOI] [PubMed] [Google Scholar]
- 6.Liljeroos MA, Miller JL, Lennie TA, Chung ML. Quality of life and family function are poorest when both patients with heart failure and their caregivers are depressed. Eur J Cardiovasc Nurs. 2022;21(3):220–226. [DOI] [PubMed] [Google Scholar]
- 7.Janson P Willeke K Zaibert L, et al. Mortality, morbidity and health-related outcomes in informal caregivers compared to non-caregivers: a systematic review. Int J Environ Res Public Health. 2022;19(10):5864. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Whittingham K, Barnes S, Gardiner C. Tools to measure quality of life and carer burden in informal carers of heart failure patients: a narrative review. Palliat Med. 2013;27:596–607. [DOI] [PubMed] [Google Scholar]
- 9.Burton AM Sautter JM Tulsky JA, et al. Burden and well-being among a diverse sample of cancer, congestive heart failure, and chronic obstructive pulmonary disease caregivers. J Pain Symptom Manage. 2012;44(3):410–420. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Dellafiore F, Chung ML, Alvaro R, Zeffiro V, Ercole V, Pucciarelli G. Influence of mutuality on quality of life in heart failure patient with inadequate self-care and caregiver dyads: an actor-partner interdependence model analysis. Eur J Cardiovasc Nurs. 2022;21(4):366–373. [DOI] [PubMed] [Google Scholar]
- 11.Bidwell JT, Lyons KS, Lee CS. Caregiver well-being and patient outcomes in heart Failure: a meta-analysis. J Cardiovasc Nurs. 2017;32(4):372–382. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Clark AM Spaling M Harkness K, et al. Determinants of effective heart failure self-care: a systematic review of patients' and caregivers' perceptions. Heart. 2014;100(9):716–721. [DOI] [PubMed] [Google Scholar]
- 13.Wang Z, Tocchi C. Partners' experience of informal caregiving for patients with heart Failure: a meta-ethnography. J Cardiovasc Nurs. 2023;38(2):E40–E54. [DOI] [PubMed] [Google Scholar]
- 14.Iavarone A, Ziello AR, Pastore F, Fasanaro AM, Poderico C. Caregiver burden and coping strategies in caregivers of patients with Alzheimer's disease. Neuropsychiatr Dis Treat. 2014;10:1407–1413. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Reblin M, Donaldson G, Ellington L, Mooney K, Caserta M, Lund D. Spouse cancer caregivers' burden and distress at entry to home hospice: the role of relationship quality. J Soc Pers Relat. 2016;33(5):666–686. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Majestic C, Eddington KM. The impact of goal adjustment and caregiver burden on psychological distress among caregivers of cancer patients. Psychooncology. 2019;28(6):1293–1300. [DOI] [PubMed] [Google Scholar]
- 17.Hellhammer DH, Wüst S, Kudielka BM. Salivary cortisol as a biomarker in stress research. Psychoneuroendocrinology. 2009;34(2):163–171. [DOI] [PubMed] [Google Scholar]
- 18.Lee DY, Kim E, Choi MH. Technical and clinical aspects of cortisol as a biochemical marker of chronic stress. BMB Rep. 2015;48(4):209–216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Weitzman ED, Fukushima D, Nogeire C, Roffwarg H, Gallagher TF, Hellman L. Twenty-four hour pattern of the episodic secretion of cortisol in normal subjects. J Clin Endocrinol Metab. 1971;33(1):14–22. [DOI] [PubMed] [Google Scholar]
- 20.Allen AP Curran EA Duggan Á, et al. A systematic review of the psychobiological burden of informal caregiving for patients with dementia: focus on cognitive and biological markers of chronic stress. Neurosci Biobehav Rev. 2017;73:123–164. [DOI] [PubMed] [Google Scholar]
- 21.Park J, Ross A, Klagholz SD, Bevans MF. The role of biomarkers in research on caregivers for cancer patients: a scoping review. Biol Res Nurs. 2018;20(3):300–311. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Bani-Issa W Radwan H Al Marzooq F, et al. Salivary cortisol, subjective stress and quality of sleep among female healthcare professionals. J Multidiscip Healthc. 2020;13:125–140. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Benfield RD, Newton ER, Tanner CJ, Heitkemper MM. Cortisol as a biomarker of stress in term human labor: physiological and methodological issues. Biol Res Nurs. 2014;16(1):64–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Jung C Greco S Nguyen HHT, et al. Plasma, salivary and urinary cortisol levels following physiological and stress doses of hydrocortisone in normal volunteers. BMC Endocr Disord. 2014;14(1):91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Novak M, Guest C. Application of a multidimensional caregiver burden inventory. Gerontologist. 1989;29(6):798–803. [DOI] [PubMed] [Google Scholar]
- 26.Greco A Pancani L Sala M, et al. Psychometric characteristics of the caregiver burden inventory in caregivers of adults with heart failure. Eur J Cardiovasc Nurs. 2017;16(6):502–510. [DOI] [PubMed] [Google Scholar]
- 27.Chou KR, Jiann-Chyun L, Chu H. The reliability and validity of the Chinese version of the Caregiver Burden Inventory. Nurs Res. 2002;51(5):324–331. [DOI] [PubMed] [Google Scholar]
- 28.Tseng H, Lu J, Tsai Y. Assessment of health-related quality of life (II): norming and validation of SF-36 Taiwan version. Taiwan J Public Health. 2003;22:512–518. [Google Scholar]
- 29.Lu J, Tseng H, Tsai Y. Assessment of health-related quality of life in Taiwan (I): development and psychometric testing of the SF-36 Taiwan version. Taiwan J Public Health. 2003;22:501–511. [Google Scholar]
- 30.Ware J. SF-36 Health Survey: Manual and Interpretation Guide. Boston, MA: Nimrod; 1993. [Google Scholar]
- 31.Pressler SJ Gradus-Pizlo I Chubinski SD, et al. Family caregiver outcomes in heart failure. Am J Crit Care. 2009;18(2):149–159. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Chung ML, Moser DK, Lennie TA, Rayens MK. The effects of depressive symptoms and anxiety on quality of life in patients with heart failure and their spouses: testing dyadic dynamics using actor-partner interdependence model. J Psychosom Res. 2009;67(1):29–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Bell CC. DSM-IV: diagnostic and statistical manual of mental disorders. JAMA. 1994;272(10):828–829. [Google Scholar]
- 34.Wang W Bian Q Zhao Y, et al. Reliability and validity of the Chinese version of the Patient Health Questionnaire (PHQ-9) in the general population. Gen Hosp Psychiatry. 2014;36:539–544. [DOI] [PubMed] [Google Scholar]
- 35.Leung DYP, Mak YW, Leung SF, Chiang VCL, Loke AY. Measurement invariances of the PHQ-9 across gender and age groups in Chinese adolescents. Asia Pac Psychiatry. 2020;12(3):e12381. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Chang ML, Tsai FC, Wang CL, Chen CC, Huang TY. Depression and related factors of patients with chronic heart failure in Taiwan. Chang Gung Nursing. 2012;23(2):145–155. [Google Scholar]
- 37.Cohen J. Statistical Power Analysis for the Behavioral Sciences (2nd ed.). 1988. Routledge: New York. 10.4324/9780203771587 [DOI] [Google Scholar]
- 38.Saunders MM. Indicators of health-related quality of life in heart failure family caregivers. J Community Health Nurs. 2009;26(4):173–182. [DOI] [PubMed] [Google Scholar]
- 39.Clements L, Frazier SK, Moser DK, Lennie TA, Chung ML. The mediator effects of depressive symptoms on the relationship between family functioning and quality of life in caregivers of patients with heart failure. Heart Lung. 2020;49(6):737–744. [DOI] [PubMed] [Google Scholar]
- 40.Swinkels J, Tilburg TV, Verbakel E, Broese van Groenou M. Explaining the gender gap in the caregiving burden of partner caregivers. J Gerontol B Psychol Sci Soc Sci. 2019;74(2):309–317. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Gu H, Ma X, Zhao J, Liu C. A meta-analysis of salivary cortisol responses in the Trier Social Stress Test to evaluate the effects of speech topics, sex, and sample size. Compr Psychoneuroendocrinol. 2022;10:100125. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Chiou CJ, Chen IP, Wang HH. The health status of family caregivers in Taiwan: an analysis of gender differences. Int J Geriatr Psychiatry. 2005;20(9):821–826. [DOI] [PubMed] [Google Scholar]
- 43.Huang CY Sousa VD Perng SJ, et al. Stressors, social support, depressive symptoms and general health status of Taiwanese caregivers of persons with stroke or Alzheimer's disease. J Clin Nurs. 2009;18(4):502–511. [DOI] [PubMed] [Google Scholar]
- 44.Hsiao CY. Family demands, social support and caregiver burden in Taiwanese family caregivers living with mental illness: the role of family caregiver gender. J Clin Nurs. 2010;19(23–24):3494–3503. [DOI] [PubMed] [Google Scholar]
- 45.Tang ST, Liu TW, Lai MS, McCorkle R. Discrepancy in the preferences of place of death between terminally ill cancer patients and their primary family caregivers in Taiwan. Soc Sci Med. 2005;61(7):1560–1566. [DOI] [PubMed] [Google Scholar]
- 46.Lin PC, Lu CM. Hip fracture: family caregivers' burden and related factors for older people in Taiwan. J Clin Nurs. 2005;14(6):719–726. [DOI] [PubMed] [Google Scholar]
- 47.Ku LE, Chang SM, Pai MC, Hsieh HM. Predictors of caregiver burden and care costs for older persons with dementia in Taiwan. Int Psychogeriatr. 2019;31(6):885–894. [DOI] [PubMed] [Google Scholar]