From psychological resilience to work engagement: a chain mediation of cognitive reappraisal and self-efficacy in nurses

1 Introduction

Under the dual challenges posed by the high workload of the healthcare system and the normalization of public health events, nurses in tertiary hospitals—as the core human resources within China's healthcare system—are confronting a steadily escalating occupational mental health crisis. At the level of the practice environment, they must cope with high workload intensity (Jia et al., 2024). At the level of technical competence, nurses in tertiary hospitals are required to master a range of specialized techniques and to undertake teaching responsibilities, such as the standardized training of new nurses and the professional development of specialized nurses. Emotional exhaustion resulting from occupational stress not only poses a direct threat to nurses' physical and mental health but also serves as a core risk factor for the quality of patient care (Ma et al., 2025).

Psychological resilience is not merely a static trait that individuals employ to cope with stress (Garmezy et al., 1984); it is also an adaptive psychological resource that can be dynamically enhanced through cognitive strategies (Luthar et al., 2000). Work engagement refers to a positive, fulfilling, and energetic state of mind that individuals exhibit in their work (Schaufeli et al., 2002). Cognitive reappraisal, as a positive emotion regulation strategy, involves the cognitive reinterpretation and re-evaluation of emotional situations or events (Gross, 1998) and has been identified as an effective intervention for managing negative emotional experiences (Gross and John, 2003). General self-efficacy, as a positive dimension of psychological capital, refers to an individual's subjective perception and evaluation of their own motivation, cognitive resources, and ability to execute the behaviors necessary to cope with anticipated situations (Bandura, 1978). It is also an important driver of adaptation and behavioral change (Lauren et al., 2016).

Previous studies have indicated that psychological resilience exerts a facilitative effect with work engagement (Liu et al., 2024). Furthermore, both cognitive reappraisal and self-efficacy have been shown to positively predict work engagement (Hu et al., 2016; Wu et al., 2024; Xanthopoulou et al., 2007). Nevertheless, several limitations persist in the existing literature. First, the effect of psychological resilience on work engagement is often simplified as a linear relationship, thereby failing to reveal the synergistic pathway between situational emotion regulation and stable motivational resources within the mechanism of psychological resilience. Second, previous cross-sectional studies have relied on traditional scales, which lack a fine-grained examination of the hierarchical transmission mechanisms among mediator variables.

Drawing upon Conservation of Resources (COR) theory (Hobfoll, 2001), individuals with greater psychological resilience are better equipped not only to protect existing resources but also to generate new ones. One such resource is the ability to employ cognitive reappraisal, a situational emotion regulation strategy that enables individuals to reinterpret and reframe stressful events. By consistently applying this strategy, resilient nurses can mitigate emotional exhaustion and conserve valuable personal resources. According to Bandura's (1978) social cognitive theory, successful performance of a task—such as effectively reappraising a stressful clinical encounter—constitutes a “mastery experience,” which is the most influential source of self-efficacy. Therefore, while cognitive reappraisal is inherently situational, its repeated and successful application over time can foster a more stable and generalized belief in one's capabilities—namely, general self-efficacy. This elevated self-efficacy, in turn, motivates individuals to invest sustained effort and persistence in their work, ultimately enhancing work engagement (Lisbona et al., 2018).

Importantly, although the reverse sequence—whereby self-efficacy facilitates greater use of cognitive reappraisal—is theoretically plausible, our proposed order is rooted in the resource-generation mechanism of COR theory. Specifically, we posit that resilience enables the acquisition of emotion regulation skills, and the successful deployment of these skills subsequently builds competence beliefs, rather than vice versa. This temporal framing aligns with intervention logic: training programs typically teach cognitive strategies before expecting changes in stable beliefs. Thus, examining the “cognitive reappraisal → general self-efficacy” pathway offers novel insights into how situational emotion regulation can be transformed into durable motivational resources.

On this basis, this study constructed a chain mediation model of “cognitive reappraisal → general self-efficacy” to explore the internal mechanism linking psychological resilience to work engagement among nurses in tertiary hospitals. Specifically, we propose the following hypotheses:

• Hypothesis 1 (H1): Psychological resilience is positively associated with work engagement.
• Hypothesis 2 (H2): Cognitive reappraisal and general self-efficacy act as separate mediators in the relationship between psychological resilience and work engagement.
• Hypothesis 3 (H3): Cognitive reappraisal and general self-efficacy serve as sequential mediators in the relationship between psychological resilience and work engagement, such that higher psychological resilience is associated with greater use of cognitive reappraisal, which in turn is associated with higher general self-efficacy, ultimately linking to higher work engagement.

2 Materials and methods

2.1 Participants

We employed a total population sampling method. All registered nurses (N = 448) working in a tertiary hospital in Shenyang, China, for at least one year were invited to participate during November 2024–January 2025, and 24 questionnaires with incomplete information were excluded, and the valid questionnaires were 424, with an effective rate of 94.64%. The sample of nurses in the tertiary hospital were all female, aged 24–48 years, with a mean age of (35.65 ± 6.07) years, and with a work experience of 4–28 years, with a mean work experience of (14.93 ± 6.17) years. Participants' professional rank, a categorical variable representing their professional rank (junior nurse, intermediate nurse, or senior nurse), were also recorded based on hospital administrative records.

2.2 Research instruments

2.2.1 Short version of the psychological resilience scale (10-item Connor-Davidson resilience scale, CD-RISC-10)

Developed by Campbell-Sills et al. (Campbell-Sills and Stein, 2007) and translated and revised by Wang et al. (2010), the 10-item scale uses a 5-point Likert scale ranging from 0 (“never”) to 4 (“always”). The total score is the sum of all 10 items, with a possible range of 0–40; higher scores indicate greater psychological resilience. The Cronbach's α for this scale in this study was 0.833.

2.2.2 Emotion regulation questionnaire (ERQ)

The ERQ was developed by Gross and John (2003) and consists of 10 items divided into two dimensions: Cognitive Reappraisal (6 items, e.g., “I control my emotions by changing the way I think about the situation I'm in”) and Expressive Suppression (4 items). This study used only the Cognitive Reappraisal subscale. The ERQ is scored on a 7-point Likert scale from 1 “strongly disagree” to 7 “strongly agree,” and the score for each dimension is the sum of the scores obtained from the items in that dimension, with higher scores indicating that the individual uses a particular emotion regulation strategy more frequently. This questionnaire was used to measure the subjects' cognitive reappraisal dimension scores, and the Cronbach's alpha for the cognitive reappraisal scale in this study was 0.894.

2.2.3 General self-efficacy scale (GSES)

The General Self-Efficacy Scale (GSES) was developed by Schwarzer et al. (1995). The Chinese version was translated and revised by Wang et al. (2001). The scale consists of 10 items rated on a 4-point Likert scale ranging from 1 (“not at all true”) to 4 (“exactly true”). The total score is the sum of all items divided by 10 (i.e., the average score per item), with a possible range of 1–4; higher scores indicate higher levels of general self-efficacy. The Cronbach's α for this scale in this study was 0.946.

2.2.4 Utrecht work engagement Scale-9 (UWES-9)

The UWES-9 was developed by Schaufeli et al. (2006) and consists of 9 items covering three dimensions: vigor, dedication, and absorption. It uses a 7-point Likert scale ranging from 0 (“never”) to 6 (“always”), with the total score being the sum of all items; higher scores indicate greater work engagement. The Cronbach's α of the scale in the current study was 0.951, and the Cronbach's α coefficients of all subdimensions were > 0.8, indicating good reliability.

2.3 Data collection procedure

First, an online questionnaire was designed by members of the research team, including the purpose, significance, methodology, precautions for completing the questionnaire, and the content of the questionnaire, and a pre-experimental completion of the questionnaire was conducted to determine the final version; second, members of the research team contacted the hospital director to explain the purpose and content of the survey, and with their consent, the link to the questionnaire and the QR code were sent to the nursing working group of each department; finally, nurses, with their informed consent, completed the questionnaire on a voluntary basis. All responses were mandatory, and each IP account had only one opportunity to respond. To ensure the confidentiality of the results, all results of this study were digitally coded and was accessible only to members of the research team.

2.4 Statistical analysis

SPSS 26.0 (IBM Corp., Armonk, NY, USA) software was used for statistical analysis. First, descriptive statistics (means and standard deviations) were calculated for all study variables, and the internal consistency reliability (Cronbach's α) was examined for each scale. Second, to assess potential common method bias, Harman's one-factor test was performed. Additionally, a Confirmatory Factor Analysis (CFA) was conducted to evaluate the discriminant validity of the four core study variables: psychological resilience, cognitive reappraisal, general self-efficacy, and work engagement. The measurement model demonstrated an acceptable fit (χ²/df = 2.86, CFI = 0.91, TLI = 0.90, RMSEA = 0.07, SRMR = 0.06), supporting the distinctiveness of the four constructs. A post-hoc power analysis using G^*Power 3.1 indicated that our sample size (N = 424) provided sufficient power (> 0.90) to detect the observed small-to-medium effects in the mediation model (Cohen's f² = 0.14, α = 0.05). Third, Pearson's correlation analysis was performed to examine the bivariate relationships among all variables, including psychological resilience, cognitive reappraisal, general self-efficacy, and work engagement. Fourth, to test the hypothesized chain mediation model, we used Hayes' PROCESS macro (Version 4.2, Model 6) for SPSS. In this model, work engagement was specified as the outcome variable (Y), psychological resilience as the independent variable (X), cognitive reappraisal as the first mediator (M1), and general self-efficacy as the second mediator (M2). Age and professional rank (professional rank: junior, intermediate, or senior nurse) were included as covariates (Z1, Z2) based on prior literature indicating that demographic factors and professional seniority may influence work engagement and emotional regulation strategies (e.g., Xanthopoulou et al., 2007; Ma et al., 2025). The significance of regression coefficients and indirect effects was tested using the percentile bootstrap method with bias correction. A total of 5,000 bootstrap resamples were drawn to generate 95% confidence intervals (CIs). An effect was considered statistically significant if the 95% CI did not include zero. The significance level was set at α = 0.05.

3 Results

3.1 Common method bias test

Since all of the data in the study were collected from self-report measures, this could lead to the problem of common method bias. The Harman one-way test was used to test for common method bias for the measures involved. The results showed that there were a total of seven factors with eigenroots greater than 1. The amount of variance explained by the first factor was 18.95%, which was less than the critical value of 40%, indicating that there was no significant common method bias in this study (Tang and Wen, 2020).

3.2 Descriptive statistics and correlation analysis

The results were analyzed using Person's correlation test and showed (see Table 1) that there was a positive moderate correlation between psychological resilience of nurses in tertiary hospitals and cognitive reappraisal, general self-efficacy, and work engagement in both variables (P < 0.05).

Table 1. Means and correlation analysis for each study variable.

^a P < 0.01. Theoretical ranges indicate possible minimum and maximum scores for each scale. For GSES, the average score per item is reported (range 1–4).

3.3 Regression analysis of psychological resilience, cognitive reappraisal, self-efficacy and work engagement

Based on the theoretical assumptions and the above statistical results, to further explore the effect of psychological resilience on work engagement, multiple hierarchical regression analyses were conducted using Model 6 of Process to test the chain mediation model (see Figure 1), with work engagement as the dependent variable, psychological resilience as the independent variable, and cognitive reappraisal and general self-efficacy as the mediator variables. The mediating effects of cognitive reappraisal and general self-efficacy between psychological resilience and work engagement were examined, controlling for age and rank.

The results of regression analysis showed (see Table 2) that psychological resilience was a significant positive predictor of work engagement (β = 0.666, t = 18.921, P < 0.001), psychological resilience was a significant positive predictor of cognitive reappraisal (β = 0.682, t = 5.087, P < 0.001), cognitive reappraisal was a significant positive predictor of work engagement (β = 0.164, t = 3.439, P < 0.001); psychological resilience was a significant positive predictor of general self-efficacy (β = 0.540, t = 13.935, P < 0.001), and general self-efficacy was a significant positive predictor of work engagement (β = 0.418, t = 7.628, P < 0.001); cognitive reappraisal was a significant positive predictor of general self-efficacy (β = 0.340, t = 8.700, P < 0.001).

Table 2. Mediation model regression analysis of cognitive reappraisal and self-efficacy between psychological resilience and work engagement.

^a P < 0.05;

^b P < 0.001.

Based on the results of the regression analysis, to further explore the mediating mechanism of cognitive reappraisal and general self-efficacy, 3 pathways were hypothesized through which psychological resilience may act on work engagement in addition to the direct effect.

• Path 1: Psychological resilience → cognitive reappraisal → work engagement
• Path 2: Psychological resilience → general self-efficacy → work engagement
• Path 3: Psychological resilience → cognitive reappraisal → general self-efficacy → work engagement.

3.4 Mediation effect test

The three-way mediation effect test was conducted using the nonparametric bootstrap method (repeated samples of 5,000 times) and the bias-corrected bootstrap method, and as shown in the mediation effect results table (see Table 3), the overall effect of psychological resilience on work engagement was significant (effect value = 0.666, SE = 0.042, P < 0.001), with a direct effect value of 0.232, 95% CI: 0.155~0.403, not including 0, the direct effect was significant; the indirect effect was 0.434, 95% CI: 0.336~0.525, indicating that cognitive appraisal and self-efficacy significantly mediated between psychological resilience and work engagement, and the indirect effect accounted for 65.17% of the total effect.

Table 3. Significance analysis of mediated effects.

The mediation effect value of cognitive reappraisal was 0.112, 95%CI: 0.049~0.175, not including 0, the mediation effect was significant, and the proportion of indirect effect of cognitive reappraisal was 16.82% (0.112/0.666). The mediation effect value of general self-efficacy was 0.226, 95%CI: 0.161~0.282, not including 0, the mediation effect was significant, and the percentage of indirect effect of general self-efficacy was 33.93% (0.226/0.666). The mediation effect of “cognitive reappraisal → self-efficacy” was 0.097, 95% CI: 0.061~0.138, not including 0, with a significant chain mediation effect. The indirect effect of “cognitive reappraisal → self-efficacy” was 13.66% (0.097/0.666).

The difference of “Path 1-Path 2” is −0.114, 95% CI: −0.200~-0.018, excluding 0, which is a significant mediating effect, indicating that the mediating effect of self-efficacy is more important than that of cognitive reappraisal; The difference of “Path 1 – Path 3” is 0.015, P = 0.287, 95% CI: −0.074~0.096, including 0, indicating that there is no significant difference between the two mediation effects; comparing the mediation effect of “Path 2 – Path 3” The difference of “Path 2 – Path 3” is 0.129, 95% CI: 0.077~0.181, not including 0. The difference of the mediation effect is significant, indicating that the mediation effect of self-efficacy is more important than the chain mediation effect. All regression assumptions (linearity, normality of residuals, homoscedasticity) were checked and met. Variance Inflation Factor (VIF) values for all predictors were below 2.5, indicating no significant multicollinearity.

4 Discussion

4.1 Psychological resilience of nurses in a tertiary hospitals in relation to cognitive reappraisal, general self-efficacy and work engagement

Results of this study showed that psychological resilience among tertiary care nurses positively correlated with cognitive reappraisal, general self-efficacy, and work engagement.

Psychological resilience was positively associated with work engagement, which is consistent with the findings of Bai et al. (2024). A higher level of psychological resilience was associated with better adjustment of physical and mental states, more effective alleviation of negative emotional effects at work, and consequently was related to higher work engagement. Psychological resilience positively predicted cognitive reappraisal and general self-efficacy, which is consistent with the findings of Jiang et al. (2021). Psychological resilience, as a core individual capacity for coping with stress, may facilitate nurses' reinterpretation of stressful events and the generation of positive cognitive reappraisal. At the same time, higher psychological resilience is associated with more adaptive responses to occupational challenges, which in turn is related to higher self-efficacy.

Cognitive reappraisal was positively correlated with general self-efficacy, which is consistent with the findings of Qian et al. (2023). This suggests that more frequent use of cognitive reappraisal is associated with higher levels of general self-efficacy. Cognitive reappraisal was a significant positive predictor of work engagement, which is similar to the findings of Wu et al. (2024). As a positive emotion regulation strategy, cognitive reappraisal is associated with positive cognitions in nurses (Wang et al., 2022). Accordingly, enhanced physical and mental resources invested in work are associated with higher levels of work engagement.

Higher general self-efficacy scores were associated with higher levels of work engagement, which is consistent with the findings of Xie and Shi (2024). Self-efficacy, as a belief, is a motivational factor that encourages individuals to invest more effort and persistence in pursuing their goals (Lisbona et al., 2018), and this is associated with higher work engagement.

4.2 Chain mediation of cognitive reappraisal and general self-efficacy between psychological resilience and work engagement among nurses in a tertiary hospital

This study further explored the mediating factors between psychological resilience and work engagement through a constructed chain mediation model to examine how psychological resilience is associated with work engagement. The results indicated that cognitive reappraisal and general self-efficacy mediated the association between psychological resilience and work engagement, both individually and as a sequential chain.

Path 1 is consistent with Hobfoll's idea of “Conservation of Resources (COR) theory ” (Hobfoll, 2001), which suggests that in the face of stress, individuals who successfully acquire beneficial resources by learning new skills not only effectively counteract resource depletion but also potentially bring about new resource gains by blocking the stressful process. In the face of stress, which is associated with maintaining higher levels of work engagement. Meanwhile, the mediating role of cognitive reappraisal supports Gross's model of emotion regulation processes (Quoidbach et al., 2015) and Folkman's research on the social functioning of emotions (Folkman and Moskowitz, 2000). Highly psychologically resilient nurses are more adept at reconstructing stressful situations through cognitive reappraisal, reducing emotional exhaustion and developing new resources, thereby maintaining higher levels of work engagement.

Path 2 is consistent with Bandura's theory of self-efficacy motivation, and the study by Qin et al. (2023) confirmed that self-efficacy is positively correlated with psychological resilience and that nurses with high levels of general self-efficacy are likely to have greater confidence in a variety of situations, which may be associated with higher work engagement. Psychological resilience is associated with higher levels of work engagement through its relationship with enhanced competence beliefs, which in turn are related to confidence in performing professional activities and tasks (Li et al., 2023).

Path 3 supports a sequential process where psychological resilience is associated with greater use of cognitive reappraisal, which in turn is associated with higher general self-efficacy, ultimately leading to higher work engagement. This finding is consistent with the “mastery experience” principle of Bandura's (1978) theory: successfully reinterpreting a stressful situation (e.g., a patient complaint as an opportunity for improvement) provides a direct mastery experience, thereby strengthening an individual's belief in their ability to handle future challenges. For example, nurses change their emotional impact at work by reconstructing the positive meaning of stressful situations (Panno et al., 2023). Psychological resilience is a “cognitive reappraisal → self-efficacy” cascade that transforms intrinsic resilience into effective emotional regulation strategies and higher self-efficacy beliefs (Dawel et al., 2024). From a practical management perspective, this “emotion → cognition” pathway suggests that interventions designed to boost work engagement should be modular. For instance, a training program could first focus on cognitive reappraisal skills (e.g., reframing negative events). Subsequently, to translate this into self-efficacy, managers could assign progressively challenging tasks (mastery experiences) or provide vicarious learning through peer success stories. This structured approach could create a virtuous cycle of “emotion management → confidence enhancement → behavioral engagement,” offering a more targeted strategy than generic well-being programs.

A particularly noteworthy finding of this study is the marked discrepancy between the direct and total effects. The total effect of psychological resilience on work engagement was β = 0.666, whereas the direct effect, after accounting for the two mediators, was substantially reduced to β = 0.232. This indicates that the association between resilience and work engagement is predominantly indirect, with the total indirect effect (β = 0.434) accounting for 65.17% of the total effect. In other words, less than 35% of the association between psychological resilience and work engagement operates through a direct pathway; the vast majority is transmitted through the sequential and independent mediating roles of cognitive reappraisal and general self-efficacy.

This pattern carries important theoretical and practical implications. Theoretically, it underscores the critical role of psychological resources—specifically, emotion regulation strategies and competence beliefs—in translating resilience into positive work outcomes. Rather than functioning as a simple bivariate relationship, resilience appears to operate through a resource generation cascade: it facilitates the use of adaptive emotion regulation strategies, which in turn build self-efficacy beliefs, ultimately promoting work engagement. This finding aligns with and extends the Conservation of Resources theory (Hobfoll, 2001) by specifying the sequential resource generation pathway.

From a practical standpoint, the predominance of indirect effects suggests that interventions designed to enhance work engagement should not focus exclusively on building resilience itself. Instead, they should target the downstream mechanisms—specifically, training in cognitive reappraisal and fostering general self-efficacy—to achieve more substantial and sustainable improvements. A modular intervention approach, for instance, could first teach nurses cognitive reappraisal skills (e.g., reframing negative clinical events), then gradually introduce progressively challenging tasks to provide mastery experiences that build self-efficacy, and finally consolidate these gains through peer modeling and vicarious learning. This structured approach could create a virtuous cycle of “emotion management → confidence enhancement → behavioral engagement,” offering a more targeted and potentially more effective strategy than generic wellbeing programmes.

5 Limitation

This study has several limitations that should be acknowledged. First, the cross-sectional design precludes any inference of causality or temporal precedence among the variables. The proposed mediation model is based on theoretical assumptions, and the associations found could be bidirectional. Future longitudinal or experimental studies are needed to establish causal directions. Second, the sample was drawn from a single tertiary hospital in one city in China and consisted entirely of female nurses. This homogeneity limits the generalizability of our findings to male nurses, nurses in other healthcare settings (e.g., secondary or primary care, private clinics), or those from different cultural backgrounds. Third, all data were collected via self-report questionnaires, which raises the potential for common method bias and social desirability bias. Although Harman's single-factor test and CFA suggested that common method bias was not a severe issue, future studies could benefit from multi-source data (e.g., supervisor-rated work engagement). Fourth, the lack of an intervention component means we cannot test the practical utility of the proposed chain. Future research could develop and test training programs targeting the “cognitive reappraisal → self-efficacy” pathway. Fifth, the conceptual overlap among the constructs—particularly between psychological resilience, general self-efficacy, and the vigor/dedication dimensions of work engagement—should be acknowledged. While our Confirmatory Factor Analysis supported discriminant validity and Harman's single-factor test indicated that common method bias was not a severe issue, some shared variance is inherent given that these constructs all pertain to positive psychological functioning. This conceptual proximity may have inflated the observed correlations to some extent. Future research could benefit from using objective measures of performance or peer/supervisor ratings of engagement to further mitigate this concern and to better disentangle these interrelated constructs.

6 Conclusion

This study reveals the synergistic mechanism between cognitive reappraisal and self-efficacy in the psychological resilience and work engagement of nurses in a tertiary hospital, and provides a theoretical basis for the construction of a mental health intervention program for nurses based on the integration of emotion and cognition. Future research can test its generalizability to different levels of healthcare institutions or different cultural backgrounds, and explore how the “emotion regulation-self-efficacy” pathway can be strengthened through modular training to optimize the mental health and professional performance of the nursing population.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Carmen D. Álvarez-Albelo, University of La Laguna, Spain

Reviewed by: Te-Feng Yeh, Central Taiwan University of Science and Technology, Taiwan

Sana Mukhtar, Khwaja Fareed University of Engineering and Information Technology (KFUEIT), Pakistan