Background
Cancer is a major global health threat, posing a significant challenge and obstacle to increasing life expectancy worldwide. According to statistics, in 2020, there were approximately 4.57 million new cancer cases and 3 million new cancer-related deaths in China [1]. Among them, lung cancer has the highest incidence and mortality rates in China, while breast cancer is the most common cancer among Chinese women [2]. The daily care of cancer patients is typically provided by family caregivers. As vital components of the social support system for cancer patients, caregivers not only bear the burden of caregiving responsibilities but also manage expectations and emotions related to the fear of losing someone important to them. This phenomenon is referred to as anticipatory grief (AG) [3]. Areia and others have noted that 25.9% of caregivers of patients with advanced cancer exhibit symptoms of complicated AG [4]. AG persists from the time of the patient’s diagnosis, manifesting as feelings of yearning and longing, an inability to accept the ultimate reality, intense preoccupation with the patient’s condition, tearfulness, sleep disturbances, anger, guilt, exhaustion, separation anxiety, and existential loneliness [5]. This significantly compromises various aspects of caregivers’ health. Given the adverse outcomes associated with AG, assisting caregivers in improving negative emotions and enhancing their quality of life is imperative.
Antecedent factors and anticipatory grief
The occurrence of AG is influenced by numerous subjective and objective factors. Stroebe proposed an integrative risk factor framework for the prediction of bereavement outcome [6], which includes interconnected factors such as the nature of stressors, interpersonal resources, intrapersonal resources, appraisal, and coping, comprehensively explaining the sources of individual differences in adapting to grief. Caregivers usually engage in difficult tasks that significantly impact their physical and mental health and spend considerable time providing care that interferes with other enjoyable activities and social relationships, resulting in a relatively heavy burden of caregiving [7]. Previous study indicates a significant correlation between caregiver burden and AG in caregivers of cancer patients [8]. Moreover, within this framework, family factors, as a form of interpersonal resources, play a crucial role in the occurrence of AG. Family resilience assists in effectively coping with various crises and stressors, promoting family recovery from adversity and acquiring more social and psychological resources [9]. As a vital component of enhancing family resilience, family communication, and support contribute to alleviating AG in caregivers [10]. Psychological capital, a positive psychological resource encompassing elements such as self-efficacy, resilience, hope, and optimism, forms the foundation for individuals’ positive mental health and adaptability [11]. Previous studies confirmed that individuals with higher psychological capital had fewer depressive symptoms [12]. Consequently, caregiver burden (as a stressor), family resilience (as interpersonal resources), and psychological capital (as intrapersonal resources) can be considered antecedent factors influencing AG.
Anzeige
The potential mediating effect of cognitive appraisal
The relationship between the three antecedent factors and AG may be mediated by other psychological factors, such as cognitive appraisal. Cognitive appraisal is the process through which individuals contemplate and evaluate the impact of stressful events on their health, including assessing the nature of the stressor and evaluating their coping abilities and strategies [13]. Cognitive appraisal encompasses three forms: challenge appraisal, harm appraisal, and threat appraisal. Challenge appraisal is considered positive, while harm and threat appraisals are deemed negative. Cognitive appraisal tendency represents the tendency to adopt specific cognitive appraisal modes when facing stressful events. Previous studies have reported the influence of these antecedent factors on caregivers’ cognitive appraisal. For instance, a study on Alzheimer’s disease caregivers found that lighter caregiver burden and more positive cognitive appraisals of the caregiving experience were associated with fewer negative emotions [14]. Moreover, individuals with stronger family functioning and psychological capital are more likely to employ positive cognitive-emotional regulation [15]. Cognitive appraisal is closely linked to emotions. The qualitative study by Griffiths et al. [16]revealed that patients who appraised stressful events as threatening experienced more negative emotions, while those appraising them as challenging experienced more positive and optimistic emotions.
The potential mediating effect of coping strategies
Another potential mediating factor that may influence AG is coping strategies. Coping refers to the conscious, purposeful, and flexible regulatory behaviors individuals adopt when facing stressful events. Individuals may employ both positive and negative coping strategies in response to stressful events. Coping tendency represents the specific coping measures individuals take when confronted with stressors. Coping strategies are associated with various factors. According to Saffari’s study, caregiver burden is correlated with positive spiritual coping strategies and negative emotions [17]. In a study of infertile women, an avoidance-focused coping style was found to mediate the relationship between family support and anxiety [18]. Moreover, three positive coping strategies—problem-, emotion-, and meaning-focused—were negatively correlated with negative emotional symptoms such as fear, depression, and irritability [19]. Although some studies suggest that these factors are related to AG, few have examined the mediating effects of these factors. Identifying such mediators may help improve caregivers’ adjustment to their role and guide healthcare systems in providing more effective interventions.
The current study
Structural equation models (SEM) are the method of choice for exploring the mechanisms of psychogenesis. Compared to the SEM traditionally constructed by the frequentist paradigm, Bayesian methods based on sampling exhibit several significant advantages. Bayesian approaches do not rely on the assumption of parameter normality and do not necessitate large sample sizes [20]. Utilizing the Markov Chain Monte Carlo (MCMC) algorithm, Bayesian methods facilitate parameter estimation by iteratively drawing a large number of samples from the posterior distribution, allowing for the construction of credible intervals for indirect effects [21]. Currently, some studies have employed Bayesian methods to test for mediating effects [22, 23]. However, to the best of our knowledge, no study has demonstrated the mediating role of cognitive appraisal and coping strategies in the relationship between caregiver burden, family resilience, psychological capital, and AG.
This study draws on Stroebe’s theory and positions caregiver burden, family resilience, and psychological capital as key factors in predicting AG at the same level. This approach facilitates a comprehensive examination of their impact on AG and better reflects the independence and interaction between these factors. Based on this, the study proposed twelve hypotheses that: (H1a) Caregiver burden is positively correlated with AG; (H1b) Cognitive appraisal tendency mediates the relationship between caregiver burden and AG; (H1c) Coping tendency mediates the relationship between caregiver burden and AG; (H1d) The chain mediating effect for cognitive appraisal tendency and coping tendency between caregiver burden and AG; (H2a) Family resilience is negatively correlated with AG; (H2b) Cognitive appraisal tendency mediates the relationship between family resilience and AG; (H2c) Coping tendency mediates the relationship between family resilience and AG; (H2d) The chain mediating effect for cognitive appraisal tendency and coping tendency between family resilience and AG; (H3a) Psychological capital is negatively correlated with AG; (H3b) Cognitive appraisal tendency mediates the relationship between psychological capital and AG; (H3c) Coping tendency mediates the relationship between psychological capital and AG; (H3d) The chain mediating effect for cognitive appraisal tendency and coping tendency between psychological capital and AG. The hypothesized model is illustrated in Supplementary Fig. 1.
Anzeige
Method
Study design and participants
This study is a descriptive study with a correlational design. Participants were recruited in the order of admission of lung and breast cancer patients using consecutive sampling, with a sampling duration of four months. The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [24]. This study was conducted from June to September 2023, and all participants were from Liaoning Cancer Hospital & Institute and The First Hospital of China Medical University in Shenyang, China. A family caregiver is a family member, relative, or friend who maintains a significant relationship with the patient and undertakes the primary caregiving role without receiving any remuneration. Inclusion criteria were as follows: (1) patients diagnosed with lung or breast cancer through clinical pathological histology; (2) assuming the primary caregiving role, as confirmed by the patient; (3) age ≥ 18 years; (4) awareness of the patient’s medical condition; (5) and clear consciousness. Exclusion criteria included: (1) cognitive or mental disorders; (2) significant organ dysfunction; (3) poor reading or language expression abilities.
The study utilized the Monte Carlo power analysis for indirect effect to determine the sample size [25]. 50 caregivers were first pre-surveyed. The results showed that the correlation coefficients between caregiver burden and cognitive appraisal tendency, coping tendency, and AG were − 0.448, -0.139, 0.545 respectively; Cognitive appraisal tendency, coping tendency, and AG were 0.353, -0.420; coping tendency and AG were − 0.168. The target power was set to be 0.9, α = 0.05, the calculation suggests that a total sample size of 210 cases is required in this study. Considering the problem of non-response and incorrect filling, the original sample was enlarged by another 20% and the final sample size for the survey was determined to be 263 cases.
Measures
Sociodemographic characteristics of participants
It was collected by a self-designed questionnaire that lists questions about gender, age, educational attainment, marital status, and duration of caregiving. Clinical data, including cancer type and TNM staging, were extracted from the patient’s medical records.
Caregiver burden inventory (CBI)
The CBI was developed by Novak et al. and is mainly used to assess caregiver burden [26]. The Chinese version of CBI was used in this study. It consists of five dimensions (time-dependent burden, developmentally constrained burden, physical burden, social burden, and emotional burden) with 24 items. Items are rated on a 5-point Likert scale, from 0 (strongly disagree) to 4 (strongly agree). The higher the total CBI score, the heavier the caregiver burden. In this study, the Cronbach’s α values for each dimension and the total scale were 0.87, 0.86, 0.92, 0.90, 0.92, and 0.95, respectively.
Family resilience questionnaire (FaRE)
This scale was developed by Faccio in 2019 based on Walsh’s family resilience model [27], and the Chinese version of the FaRE scale was used in this study. The scale consists of four dimensions: communication and cohesion, navigating social support, perceiving family coping, and religion and spirituality, with a total of 24 items. Each item is scored on a 7-point Likert scale and a total score of 24–168. Higher scores on the scale suggest a stronger family resilience. In this study, the Cronbach’s α values for each dimension and the total scale were 0.93, 0.91, 0.93, 0.91, and 0.93, respectively.
Positive psychological capital questionnaire (PPQ)
The PPQ was developed by Kuo Zhang [28]. It consists of four dimensions (self-efficacy, resilience, hope, and optimism) with 26 items, and is scored on a 7-point Likert scale ranging from 26 to 182, with 5 reverse-scored items. Higher scores represent higher levels of psychological capital. In this study, the Cronbach’s α values for each dimension and the total scale were 0.86, 0.92, 0.95, 0.90, and 0.96, respectively.
Cognitive appraisal of health scale (CAHS)
The CAHS was developed by Theresa in 1998, and is mainly used to measure cancer patients’ and their caregivers’ cognitive appraisal of patients’ health-related events [29]. It was measured by the Chinese version of CAHS. The Chinese version of the CAHS consists of four dimensions: threats, challenges, harm, and benign, with a total of 24 items. The scale is based on a 5-point Likert scale, with scores from 1(strongly disagree) to 5(strongly agree). Among these, challenges and benign are regarded as positive assessments, while threats and harm are considered negative assessments. Cognitive appraisal tendency was determined by the difference between positive and negative appraisals. This approach explores individuals’ tendencies in cognitive appraisal when facing stress, reflecting their overall cognitive appraisal tendencies in different stressful situations. In this study, the Cronbach’s α values for each dimension and the total scale were 0.88, 0.91, 0.93, 0.87, and 0.92, respectively.
Simplified coping style questionnaire (SCSQ)
The SCSQ was developed by Xie [30] in 1998 and is suitable for assessing the coping styles of various populations. It consists of two dimensions (positive coping and negative coping) with 20 items. Scoring was conducted using a four-point Likert scale that ranged from 0 (not taken) to 3(often taken). Coping tendency was determined by the difference between positive and negative coping strategies. It evaluates individuals’ overall tendency to adopt either positive or negative coping strategies when facing stressful events. Coping tendency is more stable than coping style and better predicts the specific coping strategies individuals are likely to adopt in stressful situations. In this study, the Cronbach’s α values for each dimension and the total scale were 0.92, 0.82, and 0.90.
Anticipatory grief scale (AGS)
This scale was developed by Theut in 1991 and was initially used to assess anticipatory grief in caregivers of patients with dementia, and was later used in the caregivers of hospice, patients with cancer, and critically ill patients [31]. The Chinese version of AGS was used in this study. It consists of seven dimensions: sadness, feelings of loss, anger, irritability, guilt, anxiety, and ability to complete tasks, with a total of 27 items. Each item is scored on a 5-point Likert scale and a total score of 27–135. The higher the total AGS score, the heavier anticipatory grief. In this study, the Cronbach’s α values for each dimension and the total scale were 0.84, 0.85, 0.76, 0.85, 0.75, 0.88, 0.84, and 0.91, respectively.
Data collection
The questionnaire method was used in this study. Participants were recruited through the hospital’s management information system. Before the study began, the research team discussed the design of a uniform guideline for the questionnaire, which mainly included the requirements and precautions for filling in the questionnaire, and provided uniform training for the questionnaire investigators. During the questionnaire survey, the questionnaire investigator was asked to first explain to the participants the purpose and content of the study and the confidentiality of the findings, and the caregivers voluntarily signed an informed consent form to carry out the questionnaire survey. The survey questionnaires were distributed online, and it took about 30 min to complete them. When participants had questions during the questionnaire, the questionnaire investigator was ready to answer them. If the participants could not complete the survey due to personal reasons, the completed questionnaire was considered invalid. After filling out the questionnaires, participants submitted them online. To reduce the burden of responding to the questionnaires, each participant received a five yuan reward from the online system upon submission. The research team simplified the questionnaire to make sure the questions were straightforward to understand. The questions are set in close relation to the actual experiences and needs of the caregivers to increase their willingness to participate. The investigators verified the completion of the questionnaires by the participants, excluding any errors, omissions, and invalid responses to ensure the authenticity and validity of the study data.
Ethical statement
The study was approved by the Medical Ethics Committee of China Medical University (No.2023111). Based on the Declaration of Helsinki, participants had the right to leave the study at any time. Written informed consent was obtained from all participants.
Statistical analysis
All statistical analyses and data processing in this study were completed using the IBM SPSS Statistics 26.0 (IBM Corp., USA). The frequency (%) form was used to statistically describe the baseline data information of the caregivers’ sociodemographic and patients’ disease characteristics. The normality of the continuous variables such as antecedent factors, cognitive appraisal tendency, coping tendency, and AG was tested by skewness and kurtosis. After testing, the skewness of the mentioned variables ranged from − 1.320 to 0.655, and kurtosis ranged from − 0.144 to 3.447, indicating that the continuous data could be considered to obey a normal distribution [32]. Therefore, the Mean ± SD was employed to statistically describe the scale scores separately, and the correlation coefficients among them were calculated using Pearson correlation analysis. All statistical tests were two-tailed and the significance was set at 0.05. To demonstrate whether there was a chained mediation effect of cognitive appraisal tendency and coping tendency between antecedent factors and AG, we used the Bayesian SEM with IBM SPSS AMOS version 26.0 (IBM Corp., USA) to complete the data analyses. Harman’s single-factor test was used to assess common method variance (CMV). The measurement model was validated using Bayesian confirmatory factor analysis (CFA), and model convergence was assessed using convergence statistics and iteration trace plots. Due to the limited availability of referenceable model fit indices from Bayesian SEM, this study utilized Bollen-Stine Bootstrap to measure the fit coefficients of the structural model. The model fit indices were as follows: Chi-square to degrees of freedom ratio (CMIN/DF), Hoelter’s critical N (CN), root mean square error of approximation (RMSEA), goodness-of-fit index (GFI), adjusted GFI (AGFI), normed fit index (NFI), incremental fit index (IFI), comparative fit index (CFI), and Tucker-Lewis index (TLI). The recommended values for GFI, AGFI, NFI, IFI, CFI, and TLI were set at 0.90 or higher. RMSEA would be “close to” 0.09 or lower, CN at 200 or higher, and CMIN/DF “close to” 5.00 or lower, indicating a good model fit. Path coefficients of the Bayesian SEM were estimated and corrected to refine the model. Direct effects, total indirect effects, and specific indirect effects of the structural model were measured, and mediation effects were estimated in the point estimates of posterior distribution and 95% confidence interval (CI). If the 95% CI did not contain 0, the mediation effect was considered statistically significant.
Results
Sociodemographic characteristics
Survey questionnaires were distributed to 280 potential participants, resulting in 265 valid responses and an effective response rate of 94.6%, meeting the required sample size for the study. A total of 265 caregivers’ sociodemographic and patient disease characteristics are summarized in Supplementary Table 1. The mean age of caregivers was (43.25 ± 10.93) years. The majority were females (n = 265, 55.8%), bachelor and above (n = 265, 66.0%), non-religious (n = 265, 93.2%), married (n = 265, 82.6%), working (n = 265, 69.1%), and children of the patients (n = 265, 50.6%). Regarding the patients, cancer types include breast cancer (n = 265, 61.9%) and lung cancer (n = 265, 38.1%), with cancer TNM stages distributed as follows: stage I (n = 265, 32.5%), stage II (n = 265, 41.9%), stage III (n = 265, 16.9%), and stage IV (n = 265, 9.1%). The majority of cancer patients were diagnosed within the last 6 months (n = 265, 58.5%).
Descriptive statistics and correlation analysis of the variables
The AGS was 70.88 ± 17.71, CBI was 59.62 ± 16.88, FaRE was 118.03 ± 22.57, PPQ was 129.03 ± 26.14, CAHS was 1.23 ± 2.11, and SCSQ was 0.84 ± 1.25. According to the Pearson correlation analysis (Fig. 1), AG was positively associated with caregiver burden (r = 0.72) but was negatively associated with cognitive appraisal tendency, and coping tendency (r=-0.17 to -0.42). Cognitive appraisal tendency showed a positive correlation with family resilience, psychological capital, and coping tendency (r = 0.24 to 0.35), and a negative correlation with caregiver burden (r=-0.46). Psychological capital demonstrates positive correlations with family resilience and coping tendency (r = 0.34 to 0.36).
Fig. 1
Correlation coefficients for each variable. CBI1, Time dependent burden; CBI2, Developmentally constrained burden; CBI3, Physical burden; CBI4, Social burden; CBI5, Emotional burden; FARE1, Communication and cohesion; FARE2, Navigating social support; FARE3, Perceiving family coping; FARE4, religion and spirituality; PPQ1, Self-efficacy; PPQ2, Resilience; PPQ3, Hope; PPQ4, Optimism; AGS1, Sadness; AGS2, Feelings of loss; AGS3, Anger; AGS4, Irritability; AGS5, Guilt; AGS6, Anxiety; AGS7, Ability to complete tasks; CAHS, Cognitive appraisal of health scale; SCSQ, Simplified coping style questionnaire. Red is a positive correlation; blue is a negative correlation; the darker the color the stronger the correlation; blank is not statistically significant
×
![]()
Anzeige
Reliability and validity of the measurement model
As with all self-reported data, there was a potential for CMV arising from multiple sources [33]. A Harman’s single-factor test was conducted on the 22 key variables in the hypothesized model [34]. If the first factor explained more than half of the variance, it suggested the presence of CMV [35]. Results illustrated that 6 common factors could be extracted. However, the first factor accounted for only 27.12% of the variance, indicating that CMV was not a likely contaminant of our results.
Bayesian CFA was employed to assess the internal consistency reliability, convergent validity, and discriminant validity of each construct (Table 1). The results indicated that the composite reliability (C.R.) for each construct ranged from 0.850 to 0.920, surpassing the C.R. threshold of 0.60 [36], indicating satisfactory internal consistency. However, the standardized factor loading for some manifest variables did not reach 0.60, leading to model adjustments. CBI5 (emotional burden), FARE4 (religion and spirituality), and AGS2 (feelings of loss) were excluded from the measurement model. After these modifications, the standardized factor loading for each manifest variable ranged from 0.718 to 0.915, all significant (P<0.001), providing initial evidence of convergent validity for the measurement model. Additionally, the average variance extracted (AVE) for each construct ranged from 0.654 to 0.737, exceeding the AVE threshold of 0.50 [37], indicating acceptable convergent validity. Furthermore, the estimated interrelationships among all constructs were less than the square root of the AVE for each construct, providing support for discriminant validity [38] (Table 2).
Table 1
The Bayesian CFA of measurement model
Construct | Variable | Unstandardized factor loadings | S.E. | t-value | P | Standardized factor loadings | SMC | C.R. | AVE |
|---|---|---|---|---|---|---|---|---|---|
CB | CBI1 | 1 | 0.756 | 0.572 | 0.900 | 0.695 | |||
CBI2 | 1.167 | 0.075 | 15.551 | *** | 0.914 | 0.835 | |||
CBI3 | 1.056 | 0.070 | 15.128 | *** | 0.895 | 0.801 | |||
CBI4 | 0.928 | 0.074 | 12.470 | *** | 0.756 | 0.572 | |||
FR | FARE1 | 1 | 0.846 | 0.716 | 0.850 | 0.654 | |||
FARE2 | 0.997 | 0.076 | 13.091 | *** | 0.818 | 0.669 | |||
FARE3 | 0.522 | 0.042 | 12.527 | *** | 0.760 | 0.578 | |||
PC | PPQ1 | 1 | 0.860 | 0.740 | 0.918 | 0.737 | |||
PPQ2 | 1.141 | 0.060 | 18.987 | *** | 0.893 | 0.797 | |||
PPQ3 | 0.918 | 0.053 | 17.385 | *** | 0.845 | 0.714 | |||
PPQ4 | 0.862 | 0.051 | 17.032 | *** | 0.834 | 0.696 | |||
AG | AGS1 | 1 | 0.666 | 0.444 | 0.920 | 0.660 | |||
AGS3 | 0.817 | 0.077 | 10.649 | *** | 0.718 | 0.516 | |||
AGS4 | 1.036 | 0.087 | 11.891 | *** | 0.836 | 0.699 | |||
AGS5 | 1.067 | 0.087 | 12.308 | *** | 0.864 | 0.746 | |||
AGS6 | 1.312 | 0.101 | 13.051 | *** | 0.919 | 0.845 | |||
AGS7 | 0.894 | 0.075 | 11.992 | *** | 0.841 | 0.707 |
Table 2
Discriminant validity analysis for the measurement model
AG | CB | FR | PC | |
|---|---|---|---|---|
AG | 0.812 | |||
CB | 0.084 | 0.834 | ||
FR | -0.091 | -0.048 | 0.809 | |
PC | -0.064 | -0.006 | 0.416 | 0.858 |
Anzeige
Test of the structural model
The hypothetical theoretical model, after 98,501 iterations, meets the stable condition for parameter estimation, with a convergence index of 1.0018, indicating that the model has converged. The hypothetical model fitted the data well (CMIN/DF = 1.25, CN = 211.87, RMSEA = 0.03, GFI = 0.95, AGFI = 0.93, NFI = 0.95, IFI = 0.99, CFI = 0.99, TLI = 0.96). According to the results of the path coefficient test, hypotheses H1a, H1d, H2d, H3c, and H3d were accepted, while the remaining paths were rejected, and the hypothetical model was revised (Fig. 2). Caregiver burden negatively influenced cognitive appraisal tendency (β= -0.481, SE = 0.046, 95%CI= [-0.568, -0.386]), positively influenced AG (β = 0.757, SE = 0.043, 95%CI= [0.667, 0.839]). Family resilience positively influenced cognitive appraisal tendency (β = 0.241, SE = 0.062, 95%CI= [0.120, 0.360]). Psychological capital positively influenced cognitive appraisal tendency (β = 0.189, SE = 0.059, 95%CI= [0.069, 0.301]) and coping tendency (β = 0.236, SE = 0.066, 95%CI= [0.102, 0.364]). Cognitive appraisal tendency positively influenced coping tendency (β = 0.245, SE = 0.070, 95%CI= [0.108, 0.381]). Coping tendency positively influenced AG (β=-0.103, SE = 0.050, 95%CI= [-0.201, -0.007]). The path coefficient is illustrated in Supplementary Table 2. Mediation effect results confirmed that chain mediating effect for cognitive appraisal tendency and coping tendency between caregiver burden (β = 0.009, SE = 0.005, 95%CI= [0.001, 0.021]), family resilience (β=-0.003, SE = 0.002, 95%CI= [-0.009, 0.000]), psychological capital (β=-0.002, SE = 0.001, 95%CI= [-0.004, 0.000]) and AG, respectively (Table 3). That is, caregiver burden, family resilience, and psychological capital each first affect cognitive appraisal tendency, which then affects coping tendency, and coping tendency ultimately affects AG. The indirect effects of caregiver burden, family resilience, and psychological capital on anticipatory grief show a roughly similar fluctuation range in the iteration trace plots (Supplementary Fig. 2), demonstrating that the tendencies for cognitive appraisal and coping fit well as a chained mediation in the relationship between caregiver burden, family resilience, and psychological capital on AG. Additionally, coping appraisal tendency acted as a mediator between psychological capital and AG (β=-0.009, SE = 0.005, 95%CI= [-0.020, -0.001]).
Fig. 2
The chain mediating effect of cognitive appraisal tendency and coping tendency. CBI1, Time dependent burden; CBI2, Developmentally constrained burden; CBI3, Physical burden; CBI4, Social burden; FARE1, Communication and cohesion; FARE2, Navigating social support; FARE3, Perceiving family coping; PPQ1, Self-efficacy; PPQ2, Resilience; PPQ3, Hope; PPQ4, Optimism; AGS1, Sadness; AGS3, Anger; AGS4, Irritability; AGS5, Guilt; AGS6, Anxiety; AGS7, Ability to complete tasks
×
![]()
Table 3
Total indirect, specific indirect, and direct effects in Bayesian SEM
Path | Point estimate | Standard error | Z | 95%CI | |
|---|---|---|---|---|---|
Lower | Upper | ||||
Total indirect effects | |||||
CB→AG | 0.032 | 0.022 | 1.455 | -0.010 | 0.075 |
FR→AG | -0.017 | 0.010 | -1.700 | -0.040 | 0.000 |
PC→AG | -0.015 | 0.007 | -2.143 | -0.030 | -0.003 |
Specific indirect effects | |||||
CB→CAT→AG | 0.023 | 0.022 | 1.045 | -0.019 | 0.065 |
CB→CO→AG | 0.000 | 0.006 | 0.000 | -0.013 | 0.013 |
CB→CAT→CO→AG | 0.009 | 0.005 | 1.800 | 0.001 | 0.021 |
FR→CAT→AG | -0.008 | 0.008 | -1.000 | -0.027 | 0.007 |
FR→CO→AG | -0.006 | 0.005 | -1.200 | -0.019 | 0.003 |
FR→CAT→CO→AG | -0.003 | 0.002 | -1.500 | -0.009 | 0.000 |
PC→CAT→AG | -0.004 | 0.004 | -1.000 | -0.014 | 0.003 |
PC→CO→AG | -0.009 | 0.005 | -1.800 | -0.020 | -0.001 |
PC→CAT→CO→AG | -0.002 | 0.001 | -2.000 | -0.004 | 0.000 |
Direct effects | |||||
CB→AG | 0.583 | 0.064 | 9.109 | 0.468 | 0.718 |
FR→AG | -0.001 | 0.032 | -0.031 | -0.062 | 0.062 |
PC→AG | -0.002 | 0.019 | -0.105 | -0.040 | 0.035 |
Discussion
The current study assessed the relationships among caregiver burden, family resilience, psychological capital, cognitive appraisal tendency, coping tendency, and AG. Additionally, a series of chained mediation models were employed to examine how caregiver burden, family resilience, and psychological capital influence AG through the mediating effects of cognitive appraisal tendency and coping tendency.
Anzeige
According to the Bayesian SEM, caregiver burden had a positive and direct effect on AG. This finding is consistent with a previous study which indicated that alleviating the caregiving burden among caregivers of cancer patients can mitigate AG [39]. Caregivers in the process of caring for cancer patients face multiple pressures and challenges. They had to invest a significant amount of time shouldering the responsibilities and obligations of caring for the patients, potentially resulting in a substantial burden of time dependence [40]. Moreover, while taking care of the patients, they might need to balance their work, family, and social lives, [41]potentially leading to an increased developmentally constrained burden and social burden. Additionally, caregivers might encounter uncertainty related to changes in the cancer patient’s condition and the treatment process, which could also impact the caregiver burden. These factors may directly influence the emotional and psychological states of the caregivers.
The impact of cognitive appraisal tendency on AG was not significant, indicating the absence of a mediating role of cognitive appraisal tendency between caregiver burden, family resilience, psychological capital, and AG. This supports the Stress, Appraisal, and Coping theory. According to Lazarus [42], after following primary appraisal, individuals proceed to secondary appraisal to assess their coping abilities, strategies, and resources to determine how well their coping matches the stressful event. When individuals cannot effectively match, their negative emotions such as anxiety and grief may emerge. Additionally, Folkman’s research identified that coping was strongly related to cognitive appraisal; the forms of coping that were used varied depending on what was at stake and the options for coping [13]. This implies that cognitive appraisal is not the sole mediator between stressful events and AG.
The main finding of this study confirmed the chained mediating role of cognitive appraisal tendency and coping tendency. The mediation analyses revealed the indirect effect of caregiver burden on AG. Extending previous studies, this finding demonstrates that cognitive appraisal and coping not only mediate the relationship between religious beliefs and well-being [43] but also play a mediating role between caregiver burden and AG. When caregivers experience a heavier burden, it influences their cognitive appraisal, shaping their perceptions of caregiving tasks and assessments of their coping abilities. This cognitive appraisal may, in turn, affect their coping strategies—specific behaviors and psychological responses adopted when facing the burden [44]. If individuals cannot effectively cope, AG may emerge.
Caregiver burden is prevalent in the caregiving process and has direct and indirect roles in AG. Therefore, it is recommended that healthcare professionals should carry out support and interventions that address several manifestations of caregiver burden; in addition, attention should be paid to caregivers’ cognitive appraisal and coping tendencies, and appropriate psychological support should be provided to help them face the challenges of caring for patients with cancer more effectively, to reduce their caregiving burden, to promote psychological well-being, and to enhance the quality of life and the quality of care.
Family resilience could influence AG through a fully chain-mediated pathway involving cognitive appraisal tendency and coping tendency. Following Oh [45], family resilience was a multidimensional concept encompassing six attribute characteristics: collective confidence, interconnectedness, positive life view, resourcefulness, open communication patterns, and collaborative problem-solving. Resilient families possessed confidence and cohesion, along with cognitive abilities to recognize and utilize social support and resources. Additionally, by sharing personal emotions and collaborating to solve problems, caregivers could enhance their coping capabilities, prompting them to face the various stresses and challenges in the caregiving process more optimistically [46]. This positive coping enhanced psychological adaptation, thereby alleviating AG.
Therefore, enhancing family resilience may be an important measure to prevent ag. Good family resilience is a crucial resource for caregivers in coping with stress and adjusting emotions. Families with high resilience can provide caregivers with more adequate financial, caregiving, and emotional support, helping them better cope with stress. Consequently, healthcare professionals can improve caregivers’ family resilience by encouraging support and cooperation among family members and providing relevant psychological interventions and caregiving support. This approach can better assist caregivers in effectively handling the challenges of caregiving, reducing their psychological burden, and enhancing overall mental health.
Mediation analysis confirmed that psychological capital had a significant indirect impact on AG, either independently through coping tendency or cognitive appraisal tendency and coping tendency together in a series. Psychological capital emphasizes factors such as a positive mindset, adaptability, and flexibility when individuals face stress and challenges [47]. The path of psychological capital-cognitive appraisal tendency-coping tendency-AG suggests that the accumulation of psychological capital shapes caregivers’ optimistic cognitive appraisals, leading them to perceive stressful events as challenging yet meaningful experiences rather than mere burdens. Consequently, caregivers are more inclined to adopt effective and adaptive emotional coping strategies [48], reducing the occurrence of grief emotions. Globally, caregivers of cancer patients face similar multiple pressures and challenges. Therefore, understanding the relationships between these variables and their mechanisms of influence can help develop more targeted support plans, improve caregivers’ mental health and quality of life, and enhance the quality of care for cancer patients. It is recommended that healthcare professionals provide support tailored to caregiver burden, family resilience, and psychological capital, enhancing death education for caregivers, raising awareness, and assisting them in coping with the impending loss of their loved one’s pain and sorrow.
Clinical implications
This study makes significant contributions to the clinical nursing practices for caregivers of cancer patients. The study indicated that the relationships between caregiver burden, family resilience, psychological capital, and AG are embedded within the chain mediating effects of cognitive appraisal tendency and coping tendency. This knowledge can be utilized to develop advanced support programs for caregivers of cancer patients, with a focus on individual factors contributing to AG. Such initiatives can assist them in more effectively coping with negative emotions, enhancing their quality of life, and improving the quality of care provided to cancer patients. Healthcare providers can implement various psychological interventions for caregivers of cancer patients, such as teaching mindfulness practice techniques, enhancing communication and mutual support among family members, and providing opportunities for social activities to promote peer support, thereby alleviating psychological stress. Additionally, death education can be offered to help family members cope with changes in the patient’s condition and eventual death, thereby reducing their psychological stress and grief.
Study limitations
This study has some limitations. Firstly, data collection relied on self-reports from family caregivers, introducing the possibility of CMV. However, given that the variables examined in this study reflect personal psychological states, collecting data directly from caregivers has a reasonable theoretical basis. Additionally, we conducted Harman’s single-factor test to mitigate or avoid the potential impact of CMV. Secondly, considering that the sociodemographic characteristics of the participants and the disease characteristics of the patients are not the primary outcomes of the study, no intergroup comparisons were conducted. Finally, as our study employed a cross-sectional design, caution is advised in interpreting the results of the mediation analysis. Future research is recommended to acquire longitudinal data to elucidate causal relationships.
Conclusions
The relationships between caregiver burden, family resilience, and psychological capital with AG are embedded in the mediating effects of cognitive appraisal and coping. Early identification and intervention for caregiver burden, family resilience, psychological capital, cognitive appraisal, and coping methods may prevent anticipatory grief in caregivers of cancer patients.
Acknowledgements
The authors would like to thank all the participants and hospital who agreed to participate in this study.
Declarations
Ethical approval
The study was approved by the Medical Ethics Committee of China Medical University (No.2023111). Based on the Declaration of Helsinki, participants had the right to leave the study at any time. Written informed consent was obtained from all participants.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
