The relationship between psychological resilience and caregiver burden among caregivers of patients with bipolar disorder: the mediating role of coping styles

Abstract

Background

Bipolar disorder (BD) causes ongoing stress for family caregivers due to its recurring nature. Resilience can reduce stress but may also lead to excessive coping efforts, causing chronic strain. Coping styles, reflecting resilience, can both mitigate immediate stress and worsen long-term fatigue. Current research overlooks how BD-specific factors like aggression influence the “resilience-coping-burden” relationship. This study examines how coping styles mediate these effects to guide targeted interventions.

Methods

A cross-sectional and descriptive study was conducted among 176 family caregivers of hospitalized bipolar disorder patients from Renmin Hospital of Wuhan University in China using a random sampling method. Data were collected through validated scales, including the Connor-Davidson Resilience Scale, Simplified Coping Style Questionnaire, and Zarit Burden Interview. Analysis involved SPSS and PROCESS Model 4 to assess mediation effects.

Results

The study revealed that psychological resilience positively influenced coping styles (β = 0.223) and negatively influenced caregiver burden (β=-0.189). However, coping styles were positively associated with caregiver burden (β = 0.351), demonstrating a mediating effect. The total effect of psychological resilience on caregiver burden, mediated by coping styles, was − 0.110.

Conclusions

Psychological resilience exerts dual effects on burden through coping styles - directly reducing burden while potentially increasing chronic stress via excessive active coping. Clinical interventions should integrate resilience enhancement with adaptive coping strategies.

Trial registration

Clinical trial number: not applicable.

Introduction

Bipolar Disorder (BD), as defined by ICD-11 [1], is ranked as the sixth leading cause of disability among psychiatric disorders worldwide [2], demonstrates a lifetime prevalence of 2.40% [3] and is characterized by recurrent episodes of manic and depressive states [4]. Beyond impairing patients’ occupational functioning and quality of life [5], its relapsing-remitting nature imposes sustained systemic strain on family systems—approximately 72.00% of BD patients require long-term family caregiving [6], exposing caregivers to multidimensional stressors: financial burden (23.00%-45.00% annual cost escalation), emotional distress (38.70% depression prevalence), and role conflict [7, 8]. Notably, caregivers of psychiatric patients exhibit significantly higher burden intensity compared to those caring for other chronic conditions [9], with a demonstrated dose-response relationship to deteriorated care quality [10], underscoring the urgency for targeted interventions.

Psychological resilience, a core mechanism of stress adaptation, maintains psychological homeostasis through dynamic integration of internal and external resources [11]. Its dual-compensatory model operates bidirectionally: when external supports (e.g., community service accessibility) are inadequate, high-resilience individuals activate endogenous protective factors via cognitive reframing (e.g., positive reappraisal) [12]; conversely, social networks provide compensatory buffering when intrinsic resources are deficient [13]. This dynamic equilibrium elucidates nonlinear associations between caregiving stress and burden perception—high-resilience caregivers exhibit lower burden levels (β=-0.32, p < 0.01) under equivalent stress loads, potentially mediated by differential coping strategy utilization [14].

Coping styles, as behavioral manifestations of resilience, may serve pivotal mediating roles in stress transmission. Active coping (e.g., problem-solving orientation) reduces perceived burden through enhanced controllability [15], whereas passive coping (e.g., emotional avoidance) exacerbates stress accumulation [16]. Studies have demonstrated the significant relationship between coping strategies and resilience in managing chronic conditions [17, 18], particularly highlighting their potential mediating effects in stress adaptation processes. However, their specific interaction patterns in BD caregiving context require further investigation. Although this pathway has been validated in general caregiving populations [19], BD-specific clinical features—including disease unpredictability, aggression risks, and treatment adherence fluctuations—may reconfigure stress-coping patterns, challenging the generalizability of existing models. Elucidating the “resilience-coping style-caregiving burden” pathway will not only advance theoretical frameworks for BD-specific stress buffering but also inform modularized interventions (e.g., resilience training coupled with coping skills development), offering critical public health solutions for improving quality of life among 46 million BD-affected families globally. Empirical studies have highlighted the unique challenges in BD caregiving context. Research has shown that BD caregivers face distinct stressors compared to other psychiatric caregivers, particularly due to the episodic nature of BD and management of manic symptoms [20]. Additionally, cultural comparative studies have revealed significant differences in how BD caregivers from Asian and Western contexts utilize support systems and coping strategies [21].

Despite these theoretical foundations and empirical evidence in general caregiving contexts, several critical gaps remain in BD caregiving research. First, although the relationship between psychological resilience and caregiver burden has been established in general psychiatric caregiving [14, 19], how this relationship manifests specifically in BD caregiving context remains unclear. Second, while coping styles have been identified as potential mediators in stress-adaptation processes, their precise mediating role between resilience and burden has not been systematically examined among BD caregivers. Third, most existing studies have focused on direct effects between variables, neglecting the potential differential impacts of specific resilience dimensions (optimism, strength, resilience) on caregiver burden. Additionally, the majority of research has been conducted in Western contexts, limiting our understanding of these relationships within Asian cultural settings, where family caregiving patterns and social support structures may differ significantly.

Research questions

1. Does psychological resilience in BD caregivers directly and negatively predict caregiver burden?

2. Do coping styles mediate the relationship between psychological resilience and caregiver burden in BD caregivers?

3. Do the effects of different dimensions of psychological resilience (optimism, strength, resilience) on caregiver burden exhibit differential pathways?

Study aims

This study aims to: (1) examine the relationship between psychological resilience and caregiver burden among BD caregivers; (2) investigate the mediating role of coping styles in this relationship; and (3) explore the differential effects of specific resilience dimensions on caregiver burden in the Chinese cultural context.

Methods

Study design

A descriptive cross-sectional study was designed.

Study setting and sample

This study was conducted at the Department of Psychiatry, Renmin Hospital of Wuhan University, a Grade A tertiary hospital in central China, from March 2024 to January 2025. The participants were 176 primary caregivers of patients with bipolar disorder who were hospitalized, who were diagnosed by qualified psychiatrists according to ICD-11 diagnostic criteria. The inclusion criteria of participants (primary caregivers): (1) Aged 18 years or older; (2) Providing care for at least 3 h per day and for a minimum duration of 7 days; (3) Signed informed consent form. Exclusion criteria for participants (primary caregivers): (1) Having mental disorders that could interfere with study participation; (2) Requesting monetary compensation for participation.

The sample size was determined based on statistical methodology and the number of variables. Following established research [22] guidelines, the sample size should be 5–10 times the number of variables, with an additional 10.00% to account for potential data loss. With 16 variables in this study (9 demographic items, 3 resilience dimensions, 2 coping dimensions, and 2 burden dimensions), the minimum required sample size was 176 participants (16 × 10 × 1.1).

Measures

Four instruments were used for data collection: a general information questionnaire, the Connor-Davidson Resilience Scale (CD-RISC-25), the Simplified Coping Style Questionnaire (SCSQ), and the Zarit Burden Interview (ZBI-22). The research team rigorously evaluated the content validity and applicability of these scales. To ensure data reliability, the following measures were implemented: (1) maintaining sample homogeneity by including only primary caregivers of patients with bipolar disorder hospitalized in the psychiatric department during the same period; (2) standardizing questionnaire formats and administration procedures; and (3) providing both written and verbal informed consent information to all participants. The reliability of all scales was verified using Cronbach’s alpha coefficients.

Personal information form

Demographic Questionnaire: Based on previous studies [23, 24], a 9-item demographic questionnaire was developed to collect sociodemographic characteristics of primary caregivers of patients with bipolar disorder. The questionnaire gathered information on gender, educational level, relationship with the patient, place of residence, patient’s hospitalization history, caregiver’s health status (chronic conditions), patient’s current functional status, and annual household income.

The Connor-Davidson Resilience Scale (CD-RISC-25)

Initially developed by Connor and Davidson [25], the Chinese version was translated and validated by Yu et al. [26]. As a gold-standard instrument for resilience assessment, the scale has been widely used in evaluating patients with post-traumatic anxiety, outpatient follow-up, and hospitalization [27]. The scale consists of 25 items across three dimensions: tenacity (13 items), strength (8 items), and optimism (4 items). Items are rated on a 5-point Likert scale (0–4, ranging from “not true at all” to “true nearly all the time”), with total scores ranging from 0 to 100. Higher scores indicate greater psychological resilience. The scale demonstrates good reliability with a Cronbach’s alpha of 0.91 for the total scale and 0.60–0.88 for individual dimensions [26]. In this study, the Cronbach’s α coefficient of the scale was 0.80, and the Cronbach’s α coefficients of each dimension ranged from 0.50 to 0.87.

The simplified coping style questionnaire (SCSQ)

This self-report instrument, developed by Xie et al. [28], consists of 20 items measuring two dimensions of coping styles. The positive coping dimension (12 items, items 1–12) assesses individuals’ tendency to use constructive problem-solving strategies, while the negative coping dimension (8 items, items 13–20) evaluates tendencies toward avoidant or passive problem-handling approaches. Each item is rated on a 4-point scale (0–3), with higher scores indicating greater tendency to employ the respective coping style. The scale demonstrates robust psychometric properties, with Cronbach’s alpha coefficients ranging from 0.84 to 0.90 for the total scale, 0.82 to 0.89 for the positive coping dimension, and 0.76 to 0.78 for the negative coping dimension, along with a test-retest reliability of 0.89 [28, 29]. The scale has been validated and shows good structural validity in Chinese populations [29]. In this study, the Cronbach’s α coefficient of the scale was 0.77, and the Cronbach’s α coefficients of each dimension ranged from 0.63 to 0.89.

The Zarit caregiver burden interview (ZBI-22)

Originally developed by Zarit et al. [30], this instrument is internationally recognized for assessing caregiver burden. This study employed the Chinese version, validated by Wang et al. [31], which comprises 22 items measuring two primary dimensions: personal strain (12 items) and role strain (6 items), plus four independent items (with item 22 serving as an overall burden assessment). Items are rated on a 5-point Likert scale (0–4, ranging from “never” to “nearly always”), with total scores ranging from 0 to 88. The burden levels are categorized into four grades: little or no burden (0–19), mild burden (20–39), moderate burden (40–59), and severe burden (≥ 60). The scale demonstrates good internal consistency with a Cronbach’s alpha coefficient of 0.87 [31, 32]. In this study, the Cronbach’s α coefficient of the scale was 0.87, and the Cronbach’s α coefficients of each dimension ranged from 0.81 to 0.82.

Ethical considerations

This study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of Renmin Hospital of Wuhan University. Written informed consent was obtained from all participants. The privacy and confidentiality of all participants were protected throughout the study process.

Data collection

Data were collected between March 2024 and January 2025 at the Department of Psychiatry, Renmin Hospital of Wuhan University. Using convenience sampling, electronic questionnaires were distributed face-to-face to primary caregivers of patients with bipolar disorder who met the inclusion criteria. Participants accessed the online questionnaire platform by scanning a QR code and completed the surveys independently. To ensure data quality and minimize potential bias, several measures were implemented: (1) limiting submissions to one per IP address; (2) setting a 20-minute time limit for completion; and (3) real-time review of submitted questionnaires by the research team to eliminate invalid responses (e.g., extremely short completion time or obvious response patterns); (4) providing on-site technical support and clarification through research assistants; (5) ensuring a quiet, private space for questionnaire completion; and (6) offering assistance to participants with limited digital literacy while maintaining response independence. Of the 194 questionnaires distributed, 176 valid responses were received, yielding a response rate of 90.72%, which met the required sample size [33]. To address potential challenges (such as time constraints and digital literacy limitations), additional measures included flexible scheduling of survey completion and enhanced technical support when needed.

Statistical analysis

Data were analyzed using SPSS version 19.0 and PROCESS macro version 4.2. Prior to the main analyses, data normality was assessed using the Kolmogorov-Smirnov test and visual inspection of Q-Q plots. Descriptive statistics were conducted to summarize demographic characteristics and main variables, with continuous variables presented as mean ± standard deviation (X̄±SD) and categorical variables as frequencies and percentages. Mann-Whitney U test and Kruskal-Wallis test were used to compare differences between groups. Chi-square tests were performed to examine associations between demographic variables (gender, education level, relationship with patient, etc.) and main study variables (psychological resilience, coping styles, and caregiver burden). Pearson correlation analysis was performed to examine the relationships among psychological resilience, coping styles, and caregiver burden. The mediating effect of coping styles between psychological resilience and caregiver burden was tested using PROCESS macro Model 4. The 95% confidence intervals calculated through bias-corrected bootstrap method (10000 resamples), which is robust to non-normality and appropriate for testing complex mediating relationships. Statistical significance was set at P < 0.05.

Results

Demographic characteristics

This study included 176 primary caregivers of patients with bipolar disorder. The average age of caregivers was 48.12 ± 8.58 years, They were females (75.00%, n = 132) and parents (55.70%, n = 98). Educational attainment was mainly at junior high school (30.70%) and senior high school/technical secondary school levels (29.50%). The majority of caregivers resided in urban areas (59.10%) and reported no chronic diseases (81.80%). Nearly half (47.70%) of the families had annual incomes below 50,000 RMB. Regarding the patients, most had experienced one (45.50%) or two (28.40%) hospitalizations. Additionally, 31.30% of patients showed no functional impairment, while 35.80% exhibited suicidal behavior or ideation (Table 1). Analysis of demographic variables showed significant associations with main study outcomes. Educational level was significantly associated with psychological resilience levels (P < 0.05), while annual household income was significantly related to caregiver burden scores (P < 0.05). Other demographic variables including age, gender, and relationship with patient did not show significant correlations with the main variables (P > 0.05).

Table 1.

Demographic characteristics and key variables of caregivers for patients with bipolar disorder (N = 176) using Chi-square test and Mann-Whitney U test for analysis

Item	n	%
Gender
Male	44	25.00
Female	132	75.00
Education Level
Primary School or Below	21	11.90
Junior High School	54	30.70
Senior High School or Vocational School	52	29.50
College or Above	49	27.80
Relationship with Patient
Spouse	9	5.10
Parent	98	55.70
Child	53	30.10
Sibling/Other	16	9.10
Residence
Urban	104	59.10
Town	41	23.30
Rural	31	17.60
Number of Patient Hospitalizations
1 Time	80	45.50
2 Times	50	28.40
3 Times or More	46	26.10
Health Status: Chronic Diseases
Yes (Hypertension, Diabetes, etc.)	32	18.20
No	144	81.80
Annual Household Income (CNY)
< 50,000	84	47.70
< 100,000	51	29.00
< 200,000	31	17.60
> 200,000	10	5.70
Patient’s Functional Status
No Disability	55	31.30
Mild Disability	34	19.30
Moderate Disability (Reduced Work)	41	23.30
Severe Disability (Unable to Work)	46	26.10
Suicidal Behavior/Intent
Yes	63	35.80
No	113	64.20

Descriptive analysis of psychological resilience, coping styles, and caregiver burden

Caregivers demonstrated a mean psychological resilience score of 61.90 ± 15.79 (range: 11–100), with the tenacity dimension exhibiting the highest subscale score (32.47 ± 8.82). Regarding coping mechanisms, the total score was 29.32 ± 8.71, characterized by a marked predominance of adaptive coping strategies (20.23 ± 7.31) over maladaptive coping patterns (9.10 ± 3.81). Assessment of caregiver burden yielded a total score of 23.76 ± 13.26, with personal burden (14.56 ± 7.60) substantially exceeding role-related burden (4.31 ± 4.02) (Table 2).

Table 2.

Total scores of psychological resilience, coping styles, and caregiver Burden(n = 176)

Item	Mean ± SD	Median	Min-Max
Psychological Resilience
Optimism	8.35 ± 2.75	8.00	2–16
Strength	21.09 ± 5.63	21.5	2–32
Resilience	32.47 ± 8.82	32.00	5–52
Total Score	61.90 ± 15.79	61.00	11–100
Coping Styles
Positive Coping	20.23 ± 7.31	21.00	1–36
Negative Coping	9.10 ± 3.81	9.00	0–18
Total Score	29.32 ± 8.71	30.00	3–47
Caregiver Burden
Personal Burden	14.56 ± 7.60	14.00	0–40
Role Burden	4.31 ± 4.02	3.00	0–16
Total Score	23.76 ± 13.26	22.00	0–71

x: Mean, SD: Standard deviation

Relationship between psychological resilience, coping styles, and caregiver burden

Correlation analyses revealed significant associations among the study variables. Total psychological resilience scores exhibited a strong positive correlation with adaptive coping strategies (r = 0.53, P < 0.001), whereas no significant relationship was identified with maladaptive coping patterns (r=-0.11, p = 0.118). Furthermore, total psychological resilience demonstrated a significant inverse correlation with overall caregiver burden (r=-0.16, P = 0.032). Maladaptive coping strategies were positively associated with both personal burden (r = 0.16, P = 0.034) and total caregiver burden (r = 0.21, P = 0.004). Of particular note, all dimensions of psychological resilience consistently showed significant negative correlations with role-related burden (P < 0.05) (Table 3).

Table 3.

Correlation analysis of resilience, coping styles and burden among caregivers of patients with bipolar disorder (n = 176)

Item	Positive Coping	Negative Coping	Total Coping Score	Personal Burden	Role Burden	Total Burden
Optimism	r: 0.50*** p:0.000	r: -0.04 p:0.522	r: 0.41*** p:0.000	r: -0.07 p: 0.300	r: -0.19* p: 0.010	r: -0.15* p: 0.042
Strength	r: 0.49*** p:0.000	r: -0.16* p:0.031	r:0.35*** p:0.000	r: 0.00 p: 0.939	r: -0.30*** p: 0.000	r: -0.14 p: 0.055
Hardiness	r: 0.48*** p:0.000	r: -0.09 p:0.229	r:0.37*** p:0.000	r:0.02 p: 0.743	r: -0.25** p: 0.001	r: -0.12 p: 0.103
Total Resilience Score	r: 0.53*** p:0.000	r: -0.11 p:0.118	r: 0.40*** p:0.000	r: -0.01 p:0.866	r: -0.29*** p < 0.001	r: -0.16* p:0.032
Positive Coping	-	-	-	r: 0.10 p:0.165	r: -0.00 p: 0.951	r: 0.05 p:0.499
Negative Coping	-	-	-	r: 0.16* p:0.034	r: 0.26*** p:0.000	r: 0.21** p:0.004
Total Coping Score	-	-	-	r: 0.15* p:0.037	r: 0.11 p:0.132	r: 0.13 p:0.067

r: Spearman Korelasyon Katsayısı

***P < 0.001

**P < 0.01

*P < 0.05

The mediating role of coping styles in the relationship between psychological resilience and caregiver burden

Mediation analysis was conducted using PROCESS macro (Model 4) to examine the hypothesized mediating role of coping mechanisms. Results demonstrated that psychological resilience exhibited a significant negative direct effect on caregiver burden (β= -0.189, 95%CI [-0.322, -0.055], P < 0.01). Bootstrap analysis with 10,000 resamples validated a significant indirect effect of psychological resilience on caregiver burden through coping mechanisms (β = 0.078, 95%CI [0.031, 0.129]), which accounted for a substantial proportion (70.7%) of the total effect. The total effect coefficient was − 0.110 (95%CI [-0.248, 0.028]), establishing coping mechanisms as a partial mediator in the relationship between psychological resilience and caregiver burden (Table 4; Fig. 1).

Table 4.

The mediating effect of coping styles between psychological resilience and caregiver burden using PROCESS model (n = 176)

Paths	Effect	SE	95%CI	%
			[LLCI, ULCI]
Total Effect	-0.110	0.070	[-0.248, 0.028]
(path c)
Direct Effect	-0.189	0.068	[-0.322, -0.055]
(path c’)
Indirect Effect(axb)	0.078	0.025	[0.031, 0.129]	70.7%

SE: Standard Error.CI: Confidence interval. LLCI: Lower Limit CI. ULCI: Upper Limit CI

Fig. 1.

The mediating model of coping styles between psychological resilience and caregiver burden. Note *** P < 0.001 ** P < 0.01 * P < 0.05

Discussion

The relationship between psychological resilience and caregiving burden among caregivers of patients with bipolar disorder is influenced by multiple factors, including demographic characteristics, coping strategies, and cultural context. Our study revealed that caregivers of patients with BD (diagnosed according to ICD-11 criteria) were predominantly female, parents, and from low-income families, consistent with trends observed in developing countries globally [32]. Notably, the high proportion of parents as primary caregivers reflects the prominent role of family bonds in Chinese culture, demonstrating the profound impact of “familism” values on intergenerational responsibilities [34]. However, this cultural identification may harbor risks: excessive emphasis on family obligations could intensify self-sacrificing tendencies, leading caregivers to neglect their own health needs [35]. This contradiction suggests that future interventions should establish mechanisms balancing individual well-being with family responsibilities while respecting cultural traditions. Our findings revealed distinctive caregiving patterns in the Chinese cultural context. The high proportion of parental caregivers (55.70%) and predominant female caregivers (75.00%) in our study reflects traditional Chinese family values [36]. Zhang et al. [36] demonstrated similar patterns in their study of Chinese caregivers, where family-centered caregiving was predominant. This cultural characteristic was further validated by Yu et al. [37], who found that Chinese caregivers showed higher family obligation scores, particularly among parent caregivers. These cultural factors may fundamentally affect how caregivers perceive and cope with their burden, especially considering that 47.70% of families in our study had annual incomes below 50,000 RMB. Similar family-centered caregiving patterns have been observed in other Asian countries, though with varying intensities [38]. Studies from Western contexts have reported different patterns in family-based care utilization and burden manifestation [39], suggesting the importance of considering both cultural specificities and universal caregiving challenges in intervention development.

The psychological resilience of bipolar disorder caregivers exhibits dynamic characteristics, modulated by patient symptom fluctuations, social support accessibility, and economic pressures. Although some caregivers demonstrated strong internal strength dimensions, overall resilience remained moderate, with the optimism dimension particularly weak. This phenomenon aligns with Johnson et al.‘s [40] dynamic resilience model, suggesting that resilience resources fluctuate with caregiving phases and external stressors. This dynamic nature of resilience has been further supported by recent meta-analytic evidence from non-Western countries, which demonstrates how cultural and socioeconomic factors significantly influence caregivers’ resilience patterns [41]. Clinical observations indicate that recurrent patient symptoms may accelerate psychological resource depletion, especially in contexts of limited social support or financial constraints [42]. Therefore, regular assessment of caregivers’ psychological status and early intervention are crucial, such as optimizing resilience resource utilization through cognitive behavioral therapy or alleviating isolation through peer support networks. Our study identified significant clinical correlates of caregiver burden. Specifically, caregivers of patients with severe functional disability (26.10%) and those caring for patients with suicidal behavior/intent (35.80%) reported higher burden scores. This finding aligns with Grover et al. [43], who demonstrated that patient disability levels significantly impact caregiver burden. The relationship between hospitalization frequency and increased caregiver burden in our study (28.40% with two hospitalizations, 26.10% with three or more) supports Sahoo et al.‘s [44] findings that clinical severity directly influences caregiving experiences. Notably, our results showed personal burden scores (14.56 ± 7.60) were substantially higher than role-related burden (4.31 ± 4.02), suggesting the significant personal impact of caregiving.

Regarding burden dimensions, the prominence of personal burden likely stems from the unpredictability of bipolar disorder symptoms and dual pressures of social stigma. The conflict between chronic emotional investment and self-identity reconstruction [45], combined with tensions between social role expectations and family responsibilities [46], exacerbates the cumulative burden effect. Although most caregivers tend to adopt positive coping strategies, the activation of negative coping strategies under high-stress conditions may become a significant trigger for increased burden [47]. This dynamic corresponds with the core tenets of stress-coping theory, whereby individuals maintain psychological homeostasis through alternating strategy use during prolonged stress [48]. Our study further revealed the mediating pathway of coping strategies between resilience and burden. Structural equation modeling demonstrated that coping styles serve as a crucial bridge between resilience and burden. This finding is particularly relevant in the context of recent meta-analytic evidence from non-Western countries, which highlights the significant role of coping strategies in mediating caregiver burden across different cultural settings [49].This finding supports Wilson et al.‘s [50] “resilience-coping-adaptation” cascade model, indicating that resilience not only directly buffers stress but also indirectly alleviates burden through positive coping strategy formation. The positive correlation between resilience and active coping suggests that highly resilient individuals tend to adopt problem-oriented strategies [51]. This mechanism aligns with resource conservation theory: resilience resources can promote active stress management through enhanced cognitive restructuring abilities [52]. Regarding this substantial mediating effect (70.7%), several methodological considerations warrant discussion. Although exceeding the mean mediation range (40–60%) in heterogeneous caregiver populations [53], comparable magnitudes have been documented specifically in populations facing chronic unpredictable stressors, as demonstrated in neurodegenerative caregivers where coping mediated 69.80% of the resilience-burden association [54]. While this finding aligns with theoretical frameworks suggesting coping as a key mechanism in the resilience-adaptation pathway, the magnitude of this effect requires careful interpretation within the context of our study design. First, the absence of certain potential confounding variables (such as social support, caregiving duration, and patient symptom severity) may influence the estimation of the mediation effect, as these unmeasured variables could share variance with both mediator and outcome variables. Second, the cross-sectional nature of data collection might have contributed to common method bias, which may inflate mediator-outcome associations through systematic measurement error. Nevertheless, our findings provide empirical support for the theoretical pathway linking resilience to caregiver burden through coping styles, while acknowledging the need for more robust methodological approaches in future research.

Based on these findings, clinical practice should transition toward multidimensional intervention modes. Healthcare institutions should integrate resilience assessment with coping strategy screening, designing stepped support programs for high-risk groups. Policy-level initiatives should promote cross-sector collaboration, incorporating respite care services and economic subsidies into regular support systems. Furthermore, public education should transcend cultural myths, advocating for shared caregiving responsibilities through destigmatizing media narratives. Future research could explore culture-specific protective factors in stress adaptation, providing differentiated intervention evidence for global mental health care. Based on our empirical findings and recent evidence [36, 43], we propose the following specific recommendations: For clinical practice:

• Regular assessment of caregiver burden using culturally validated tools [37], particularly for the 26.10% of caregivers managing severe functional disability cases.

• Implementation of targeted support programs based on patient clinical severity levels [43], especially for the 35.80% of caregivers managing patients with suicidal behavior.

• Development of culturally appropriate psychoeducation interventions focusing on enhancing psychological resilience (mean score 61.90 ± 15.79 in our study).

For healthcare services:

• Integration of caregiver support services, particularly for parent caregivers (55.70%) who form the majority of care providers [36].

• Establishment of structured support programs considering socioeconomic factors, given that 47.70% of families have low annual incomes.

• Enhanced crisis support for caregivers managing high-risk patients, especially those with multiple hospitalizations (54.50%).

Study limitations and strengths

This study makes a significant contribution by empirically validating the mediating mechanism through which psychological resilience influences caregiver burden via coping styles among caregivers of patients with bipolar disorder, providing support for the “resilience-coping-adaptation” theoretical pathway within the Chinese cultural context. However, several significant methodological limitations warrant careful consideration. First, the cross-sectional nature of our study fundamentally constrains causal inference capabilities, affecting our ability to establish temporal precedence between psychological resilience development and coping strategy adoption, observe dynamic fluctuations in burden levels corresponding to patient symptom variations, and evaluate the long-term effectiveness of different coping mechanisms. Second, our single-center convenience sampling methodology introduces potential selection bias and limits external validity. This sampling approach potentially over-represents urban residents with superior healthcare accessibility while excluding rural caregivers facing distinct resource-related challenges. The demographic characteristics of our sample may not fully reflect the diverse caregiver population in China, particularly those from different socioeconomic backgrounds and healthcare accessibility contexts. Moreover, the exclusive focus on caregivers of hospitalized patients may limit the generalizability of findings to those managing bipolar disorder in outpatient or community settings, where caregiving experiences and burden profiles are likely to differ significantly. Furthermore, our analysis was limited by the lack of control for important confounding variables such as social support, patient symptom severity, and caregiving duration, which might influence the observed relationships. Third, while our quantitative approach validated the mediating effect, it cannot fully elucidate the complex decision-making processes in coping strategy selection or capture the lived experiences of caregivers. Fourth, our analysis did not adequately account for several critical confounding variables that could substantially influence the resilience-burden relationship: (1) social support levels, which may buffer caregiving stress and enhance resilience resources; (2) caregiving duration, which could affect both coping strategy selection and burden accumulation; and (3) patient symptom severity, which may moderate the effectiveness of different coping mechanisms. The absence of these variables limits our ability to fully understand the complex dynamics of the resilience-burden pathway. These factors might substantially affect the resilience-burden pathway, as social support could buffer caregiving stress (as demonstrated by recent meta-analytic evidence [55]), prolonged caregiving duration might lead to resource depletion, and patient symptom severity could moderate the effectiveness of coping strategies. Although we used ICD-11 diagnostic criteria for BD, which enhances diagnostic reliability, future studies might benefit from incorporating multiple diagnostic systems for comparison.

To address these limitations, we propose implementing prospective cohort studies spanning at least 24 months with assessment points aligned with bipolar disorder episodes to track temporal sequences of resilience development and burden trajectories, conducting multi-center research incorporating diverse healthcare settings (e.g., inpatient, outpatient, community), geographical regions, and socioeconomic contexts, incorporating key confounding variables as covariates in statistical analyses, and adopting mixed-methods approaches to provide deeper insights into caregiver experiences. Particularly, hierarchical linear modeling approaches could help disentangle the complex interactions between caregiver characteristics, patient-related factors, and social support systems. These methodological improvements would strengthen the evidence base for developing targeted, culturally sensitive interventions in psychiatric nursing practice, ultimately improving care outcomes, particularly considering the unique challenges faced by caregivers in different healthcare accessibility contexts.

Conclusions

This empirical study advances both theoretical understanding and practical applications in caregiver research. The findings validate theoretical frameworks of stress-coping and resilience-adaptation processes while demonstrating their practical implications in the Chinese cultural context. Specifically, this study elucidated the complex relationship between psychological resilience and caregiver burden among caregivers of patients with bipolar disorder, validating the crucial mediating role of coping styles. The findings demonstrated that psychological resilience not only directly alleviates caregiver burden but also provides indirect protective effects through enhanced positive coping strategies. Based on these findings, we recommend establishing a stratified caregiver assessment system incorporating resilience and coping styles as routine screening indicators in clinical practice; strengthening professional training and establishing interdisciplinary collaboration models in nursing management; and improving caregiver support mechanisms with emphasis on rural service system development at the policy level. Specifically, we propose constructing an integrated “hospital-community-family” support network, implementing differentiated interventions based on educational levels and urban-rural differences, and incorporating support measures such as respite care into the healthcare insurance system. This study provides empirical evidence for optimizing psychological health support systems for caregivers of patients with bipolar disorder, supporting the concept of “evidence-driven personalized care,” with significant practical implications for enhancing family caregiving quality.

Author contributions

YFL and CHS collected and analyzed the data, and wrote the manuscript. XFL and ZYW designed the study and revised the manuscript. YS, ZZ, YH, and CS participated in the questionnaire collection. All authors read and approved the final manuscript.

Funding

No funding was received to conduct this research.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Renmin Hospital of Wuhan University, and written informed consent was obtained from all participants after they were informed about the study purpose, methods, voluntary participation, right to withdraw, and confidentiality protection.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.