The Effect of Negative Workplace Gossip on Nurses' Turnover Intentions: The Mediating Role of Anxiety

ABSTRACT

Aim

This study aims to investigate the relationship between negative workplace gossip targeting individual nurses and their turnover intentions, with a specific focus on the mediating role of anxiety.

Design

This cross‐sectional study applies affective events theory to examine how negative gossip, as an emotionally charged event, may be significantly associated with anxiety and higher turnover intentions.

Methods

Data were collected from 320 nurses working in tertiary and secondary hospitals in Sichuan Province, China. Participants completed established measures assessing negative workplace gossip, anxiety and turnover intentions. Hierarchical regression analysis, bootstrapping method and supplementary structural equation modelling were used to test the hypothesised relationships and mediation effects.

Results

Negative workplace gossip was significantly associated with turnover intentions of targeted nurses both directly and through anxiety as a mediator. Nurses who experienced higher levels of negative gossip towards them reported increased anxiety, which was positively related to their intentions to leave.

Conclusion

Negative workplace gossip is positively associated with anxiety and turnover intentions among targeted nurses. This study highlights the importance of addressing relational stressors within healthcare workplaces to support retention and mitigate turnover.

Patient or Public Contribution

Participating nurses contributed by completing self‐reported questionnaires.

Abbreviations

AET

affective events theory

ANX

anxiety

AVE

average variance extracted

CFA

confirmatory factor analysis

CI

confidence interval

CR

composite reliability

GAD‐7

generalised anxiety disorder 7‐item scale

HTMT

Heterotrait–Monotrait ratio

JD‐R

job demands‐resources theory

NWG

negative workplace gossip

SEM

structural equation modelling

STROBE

strengthening the reporting of observational studies in epidemiology

TOI

turnover intentions

VIF

variance inflation factor

1. Introduction

Nurses' turnover intentions have substantial implications for the healthcare sector, where high turnover disrupts continuity of care, increases operational costs and adds strain on remaining staff, ultimately affecting patient outcomes and overall healthcare quality (Al Sabei et al. 2020; Arredondo et al. 2024; Yi et al. 2024; Li and Jones 2013). Extensive research has examined the factors contributing to nurses' turnover intentions, emphasising aspects such as workload, job dissatisfaction, emotional exhaustion and insufficient organisational support (Liu et al. 2018; Maghsoud et al. 2022; Holland et al. 2019; Labrague and de Los Santos 2021). For example, Job Demands‐Resources Theory (JD‐R) illustrates how excessive job demands, including long hours and emotional labour, can lead to burnout, increasing nurses' inclination to consider leaving their positions (Özkan 2022; Mirzaei et al. 2021). Similarly, lack of organisational support from management or coworkers exacerbates stress and reduces job satisfaction, which further drives turnover intentions (Li et al. 2020; Pahlevan Sharif et al. 2023). However, while these studies have contributed significantly to understanding nursing turnover, less attention has been paid to negative workplace dynamics. In particular, limited research has examined how negative workplace gossip targeting individual nurses influences their turnover intentions. Nursing is highly team‐oriented and trust and respect are critical to patient safety and care quality. Therefore, understanding how targeted negative gossip influences nurses' turnover intentions is essential for addressing retention challenges (El‐Jardali et al. 2009; Marufu et al. 2021). This study thus addresses a critical gap in the literature, bringing novel insights to the underexplored role of interpersonal dynamics in influencing turnover intentions among nurses.

Negative workplace gossip, defined as the spread of unfavourable or unverified information about a colleague, is significantly associated with workplace relationships and the psychological well‐being of its targets (Sun et al. 2023; Gao et al. 2024; Zhou et al. 2021). Research indicates that gossip often fosters distrust and exclusion, deteriorating team cohesion and damaging the reputation and sense of belonging of the target (Xie et al. 2024; Martinescu et al. 2021). In the nursing context, where supportive social bonds are crucial, negative gossip targeting specific nurses is associated with reduced interpersonal trust and increased stress and emotional strain, disrupting the collaborative environment essential for effective patient care (Kim et al. 2021). Consequently, negative gossip may influence targeted nurses' turnover intentions by creating a hostile atmosphere that pushes them to distance themselves from such an environment. Prior research has shown that workplace gossip affects general life satisfaction and emotional exhaustion (Xie et al. 2024; Liu et al. 2020). However, less is known about the specific emotional mechanisms through which negative gossip influences turnover intentions, particularly anxiety among targeted nurses. Addressing this gap is crucial, as identifying these mechanisms could reveal actionable insights for fostering a healthier work environment and improving nurse retention. More specifically, limited empirical attention has been given to how negative workplace gossip targeting individual nurses contributes to turnover intentions through anxiety, despite the importance of interpersonal trust, communication and psychological safety in nursing teams.

To explore these relationships, this study applies affective events theory (AET), which posits that specific emotionally charged events at work elicit affective responses, shaping employees' attitudes and behaviours (Weiss and Cropanzano 1996). According to AET, highly emotional events, such as negative gossip, may trigger strong affective reactions. For nurses targeted by negative gossip, their concerns about their reputation and job security may be heightened. This uncertainty regarding colleagues' and supervisors' perceptions may be associated with a reduced sense of safety and acceptance among nurses. This, in turn, was concurrently related to increased anxiety. Moreover, the fear of potential consequences from the gossip, such as social isolation or professional setbacks, was further associated with elevated stress levels. Anxiety, a common emotional response in stressful work environments, is associated with decreased job satisfaction, impaired cognitive functioning and heightened turnover intentions as individuals seek to escape hostile or psychologically unsafe conditions (Li et al. 2024; Zalewska 2011). Despite growing interest in workplace emotions, the application of AET to understand how negative workplace gossip triggers anxiety and shapes nurses' turnover intentions remains largely unexplored. This lack of empirical evidence constitutes a clear gap that the present study seeks to address. This study hypothesises that negative workplace gossip directed at nurses is positively related to their anxiety. It also proposes that anxiety mediates the relationship between negative gossip and turnover intentions, as affected nurses may consider leaving their jobs to protect their emotional well‐being.

The implications of anxiety and turnover among nurses extend beyond individual experiences to impact healthcare institutions and patient care quality. High turnover rates among nurses increase the financial and operational burdens of recruitment and training and disrupt team dynamics. Frequent staff changes can also compromise patient care quality by leading to fragmented communication and reduced cohesion among caregivers (Hayes et al. 2012; Woodward and Willgerodt 2022). Additionally, the emotional burden on remaining staff members may create a negative feedback loop. As nurses compensate for departing colleagues, elevated stress and dissatisfaction may further increase turnover intentions within the team (Gedik et al. 2023). Thus, this study has practical importance. It aims to identify pathways that contribute to turnover intentions and to provide insights for healthcare administrators seeking to develop interventions that foster a supportive workplace culture. By focusing on the role of negative interpersonal behaviours, this research deepens our understanding of the emotional influences on turnover intentions and highlights strategies to enhance nurse retention and improve healthcare delivery outcomes.

In summary, this study addresses a critical gap by investigating the relationship between negative workplace gossip targeted at nurses and their turnover intentions, with anxiety as a mediating factor. By focusing on the mechanisms in the affective process triggered by negative gossip, this research underscores the importance of addressing negative interpersonal behaviours in healthcare workplaces to foster a positive, stable work environment for nurses. This study not only advances theoretical insights into the affective pathways influencing turnover intentions but also provides practical guidance for healthcare administrators to mitigate the adverse effects of negative workplace gossip, ultimately enhancing nurse retention and contributing to more resilient healthcare systems. Based on this analysis, the following hypotheses were proposed (Figure 1):

Hypothesis 1

Negative workplace gossip is positively related to turnover intentions of targeted nurses.

Hypothesis 2

Negative workplace gossip is positively related to anxiety of targeted nurses.

Hypothesis 3

Anxiety mediates the relationship between negative workplace gossip and turnover intentions of targeted nurses.

FIGURE 1.

Conceptual model of this paper.

2. Materials and Methods

2.1. Study Design

This study utilises a cross‐sectional design to examine the relationships between negative workplace gossip, anxiety and turnover intentions among nurses. A cross‐sectional approach allows for capturing data at a single point in time, enabling a snapshot of the associations among these variables within a specific healthcare setting. This study was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline (von Elm et al. 2007). The completed STROBE checklist is provided as Supporting Information S1.

2.2. Participants

The data collection took place from March to April 2022. A convenience sampling method was used to recruit nurses from tertiary and secondary hospitals across Sichuan Province. A total of 343 nurses employed in these hospitals were invited to participate. All participants had at least 1 year of nursing experience. To be eligible, participants needed to meet the following criteria: (a) be a registered nurse; (b) comprehend the study's objectives; (c) be willing and able to participate; and (d) currently work in their assigned ward. The survey link was shared with participants via the WeChat social media platform. Of the 343 questionnaires distributed, 325 (94.8%) were returned completed. Following the exclusion of five questionnaires due to incomplete or inconsistent responses, a total of 320 (93.3%) responses were included in the final analysis.

2.3. Measures

2.3.1. Demographic Characteristics

Demographic information, such as gender, age, education level, marital status, income, hospital tier, years of experience in the hospital and job rank, was gathered. The three primary variables of the study—negative workplace gossip, anxiety and turnover intentions—were assessed using the instruments outlined below. As the original scales were in English, we applied translation and back‐translation procedures between English and Chinese to ensure content validity. Prior to data collection, all measurement items were independently reviewed by three experts in organisational behaviour and nursing management to ensure content validity.

2.3.2. Negative Workplace Gossip

Negative workplace gossip was measured using a three‐item scale adapted from Chandra and Robinson (2009). A sample item is ‘In the past six months, others (e.g., coworkers and/or supervisors) communicated damaging information about me in the workplace’. Responses were recorded on a 5‐point Likert scale ranging from 1 (never) to 5 (always). The score for this variable is the average of all item responses. Higher scores indicated greater levels of negative workplace gossip. The Cronbach's alpha value for this measure was 0.94. Supporting the use of workplace gossip measures in Chinese organisational contexts, prior research has validated the Chinese adaptation of the Negative Workplace Gossip Scale and confirmed its reliability and factor structure (Xie et al. 2024).

2.3.3. Anxiety

Anxiety was measured using the seven‐item Generalised Anxiety Disorder scale (GAD‐7; Spitzer et al. 2006), a widely used instrument for assessing anxiety symptoms. A sample item is ‘Feeling nervous, anxious, or on edge’. In the present study, we adopted the GAD‐7 items but used a 5‐point response format (1 = not at all, 5 = nearly every day) to align with the response format of the other scales in this study. Responses were averaged to form the final score, with higher scores indicating greater anxiety. The scale demonstrated excellent reliability in this study (Cronbach's α = 0.97). While we modified the response format to 5 points for consistency with other scales in this study, the item content remained identical to the validated GAD‐7. The excellent internal consistency observed in our sample (Cronbach's α = 0.97) and the strong factor loadings support the reliability and structural validity of this adapted format in the present context. Nonetheless, we acknowledge that future research should formally validate the psychometric properties of the 5‐point GAD‐7 format in Chinese nursing populations.

2.3.4. Turnover Intentions of Nurses

A six‐item scale from Michaels and Spector (1982) was utilised to measure turnover intentions of nurses. Questions included ‘I would leave my job if I had a better opportunity elsewhere’. Responses were recorded on a 5‐point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). The score for this variable is the average of all item responses. Higher scores indicated greater levels of intentions to leave. In this study, the Cronbach's alpha value for this measure was 0.95. Psychometric validation studies conducted among Chinese healthcare workers have also confirmed the reliability and factorial validity of Chinese versions of turnover intention measures (Hu et al. 2022).

2.4. Ethical Considerations

This study was approved by the Institutional Review Board of West China Hospital, Sichuan University (Approval No: 2022572). This study was conducted in accordance with the Declaration of Helsinki. Participants were informed that their participation in this study was voluntary and they could withdraw from the survey at any time. We assured participants that the data collected would be used only for research purposes and that their information would be kept anonymous and confidential. Written informed consent was obtained prior to data collection.

2.5. Statistical Analysis

Data were analysed using IBM SPSS Version 26.0 and the PROCESS macro Version 3.5 for SPSS. Descriptive statistics were computed to summarise the demographic characteristics of the participants. Confirmatory factor analysis (CFA) was conducted to assess the validity of the measurement model. CFA is a statistical technique used to evaluate the extent to which observed variables adequately represent their underlying latent constructs (Brown and Moore 2012). Pearson's correlation analysis was then performed to examine the bivariate relationships among the study variables. To test the hypothesised relationships, hierarchical regression analysis was employed. Control variables were entered first to account for their potential confounding effects, followed by the independent variable and the mediator to examine their individual and combined effects on the dependent variable. To further test the mediating effect, bootstrapping analysis was conducted using the PROCESS macro (Hayes 2013). A total of 5000 bootstrap samples were generated to estimate the indirect effect and its 95% confidence interval, providing a robust test of mediation without relying strictly on normality assumptions. Prior to the regression analyses, key assumptions were assessed, including normality of residuals, homoscedasticity and multicollinearity. Variance inflation factor (VIF) and tolerance values were examined to assess multicollinearity. Residual diagnostics indicated that the residuals were approximately normally distributed and that no serious heteroscedasticity was detected.

3. Results

3.1. Common Method Bias

The results of Harman's single‐factor test showed that the primary factor explained 33.16% of the total variance, which was less than the critical criterion of 40% (Podsakoff et al. 2003). Therefore, this study did not present severe common method bias. In addition to Harman's single‐factor test, we acknowledge that common method variance may still inflate observed relationships. We addressed this concern through procedural remedies during data collection (Podsakoff et al. 2003): (1) ensuring respondent anonymity to reduce evaluation apprehension and (2) using established validated scales with varying response formats across constructs. While these measures do not eliminate CMV entirely, they reduce its likelihood.

3.2. Participant Characteristics

Table 1 presents the demographic characteristics of the 320 participants in this study. Among the 320 participants, the majority were female (94.7%) and only 5.3% were male. Nearly half of the nurses were under 30 years old (45.6%), 47.5% were aged 31–40 years and 6.9% were 41 years or older. Most participants were married (68.7%) and 28.8% were single. A large proportion (81.9%) held at least a bachelor's degree and 82.8% reported an annual income between ¥50,000 and ¥200,000. Regarding work experience, 39.7% had worked for 5 years or less, 36.3% had worked between 5 and 10 years and 24.0% had worked for more than 10 years. In terms of job rank, 97.5% were general nurses and the remainder held charge nurse positions.

TABLE 1.

Participant demographic variables (N = 320).

Variables	n (%)
Gender
Male	17 (5.3)
Female	303 (94.7)
Age (years)
≤ 30	146 (45.6)
31–40	152 (47.5)
41–50	17 (5.3)
≥ 51	5 (1.6)
Marital status
Married	220 (68.7)
Single	92 (28.8)
Divorced	8 (2.5)
Educational level
Low (college education and below)	58 (18.1)
High (bachelor's degree and above)	262 (81.9)
Annual income
Low (< ¥50,000)	44 (13.8)
Middle (¥50,000–¥200,000)	265 (82.8)
High (> ¥200,000)	11 (3.4)
Hospital level
Secondary	63 (19.7)
Tertiary	257 (80.3)
Years in practice
Low (< 5)	127 (39.7)
Middle (5–10)	116 (36.3)
High (> 10)	77 (24)
Nursing position
General nurse	312 (97.5)
Charge nurse	8 (2.5)

3.3. Measurement Model Validity

To assess the validity of the measurement model, we conducted confirmatory factor analysis (CFA) using IBM SPSS AMOS 26.0. Five competing measurement models were examined, including the hypothesised three‐factor model, three alternative two‐factor models and a one‐factor model. Discriminant validity was evaluated by comparing the fit indices of the hypothesised model with those of the alternative models, following Cheung and Rensvold (2002). The results showed that the three‐factor model demonstrated a substantially better fit to the data (χ ² = 207.21, df = 71, χ ²/df = 2.92, IFI = 0.98, CFI = 0.98, TLI = 0.97, RMSEA = 0.07) than the alternative models. This finding is consistent with the theoretical expectation that negative workplace gossip, anxiety and turnover intentions represent three distinct constructs.

In addition, the standardised factor loadings for all items were satisfactory, ranging from 0.82 to 0.93. As shown in Table 2, the composite reliability (CR) values for all three constructs ranged from 0.94 to 0.97, exceeding the recommended threshold of 0.70 and indicating good internal consistency. The average variance extracted (AVE) values ranged from 0.77 to 0.83, all above the recommended cutoff of 0.50, supporting convergent validity. The CFA measurement model with standardised factor loadings is presented in Figure 2. Furthermore, as presented in Table 3, the heterotrait–monotrait ratio (HTMT) values between constructs ranged from 0.73 to 0.84, all below the recommended threshold, providing additional evidence of discriminant validity. Taken together, these results indicate that the three focal variables in this study demonstrate adequate reliability as well as satisfactory convergent and discriminant validity.

TABLE 2.

Reliability and convergent validity of the study constructs.

Construct	Item	Standardised loading	CR	AVE
Negative workplace gossip	NWG1	0.90	0.94	0.83
	NWG2	0.93
	NWG3	0.89
Anxiety	ANX1	0.86	0.97	0.79
	ANX2	0.90
	ANX3	0.91
	ANX4	0.88
	ANX5	0.87
	ANX6	0.90
	ANX7	0.92
Turnover intentions	TOI1	0.82	0.95	0.77
	TOI2	0.88
	TOI3	0.91
	TOI4	0.89
	TOI5	0.86
	TOI6	0.87

Abbreviations: AVE, average variance extracted; CR, composite reliability.

FIGURE 2.

Confirmatory factor analysis measurement model of the study constructs. Standardised factor loadings are shown on the paths from latent constructs to observed items. Curved double‐headed arrows indicate correlations among the latent constructs.

TABLE 3.

HTMT matrix for discriminant validity.

Constructs	Negative workplace gossip	Anxiety	Turnover intentions
Negative workplace gossip	—	0.73	0.75
Anxiety	0.73	—	0.84
Turnover intentions	0.75	0.84	—

Note: The dash (—) on the diagonal indicates that the HTMT value of a construct with itself is not applicable.

Abbreviation: HTMT, heterotrait–monotrait ratio of correlations.

3.4. Correlation Analysis

Table 4 presents the average values, standard deviations and correlation coefficients for all variables under investigation. Negative workplace gossip was found to be significantly and positively correlated with anxiety (r = 0.64, p < 0.01) and turnover intentions (r = 0.66, p < 0.01). This indicates that nurses who are targets of negative gossip are more likely to experience heightened anxiety, which, in turn, may contribute to their intentions to leave the organisation. Furthermore, a positive and significant correlation was observed between anxiety and turnover intentions (r = 0.79, p < 0.01). This suggests that higher levels of anxiety among nurses are associated with a greater likelihood of considering turnover, likely as a means to protect their well‐being from a hostile work environment. These results offer preliminary support for the hypotheses to be tested, suggesting that the main variables in this study are interconnected in a positive manner. For ease of visual interpretation, the correlations among the key study variables are also presented in Figure 3.

TABLE 4.

Means, standard deviations (SD) and correlation coefficients of key study variables (N = 320).

Variables	Mean	SD	Negative workplace gossip	Anxiety	Turnover intentions
Negative workplace gossip	4.08	0.74	1.00
Anxiety	4.21	0.62	0.64**	1.00
Turnover intentions	4.11	0.73	0.66**	0.79**	1.00

^** p < 0.01.

FIGURE 3.

Correlation plot of key study variables. The colour intensity and values within each circle represent Pearson correlation coefficients (r). Values approaching +1 indicate a stronger positive linear relationship; values approaching −1 indicate a stronger negative linear relationship; and values near 0 indicate no linear relationship. All correlations in this figure were positive and statistically significant at p < 0.01, ranging from 0.64 to 0.79.

3.5. Hypothesis Testing

Table 5 presents the results of the hierarchical regression analyses used to test the proposed hypotheses. Gender, age and income were controlled in all models. Model 2 shows that negative workplace gossip was significantly and positively associated with anxiety (β = 0.65, p < 0.001), supporting Hypothesis 2. Model 3 further indicates that negative workplace gossip was significantly and positively associated with turnover intentions (β = 0.67, p < 0.001), supporting Hypothesis 1.

TABLE 5.

The hierarchical regression analysis (N = 320).

Variables	Anxiety		Turnover intentions
	Model 1	Model 2	Model 3	Model 4
	β	β	β	β
Gender	0.03	0.03	−0.13**	−0.13**
Age	0.02	0.02	0.07	0.06
Income	0.11	0.11	−0.07	−0.05
Negative workplace gossip		0.65***	0.67***	0.27***
Anxiety				0.62***
R 2	0.01	0.41	0.45	0.68
Adjusted R 2	0.004	0.40	0.44	0.67
F	1.42	54.98***	64.54***	131.84***
95% CI for key regression coefficients:
Negative workplace gossip → Anxiety (Model 2): [0.57, 0.73]
Negative workplace gossip → Turnover intentions (Model 3): [0.59, 0.75]
Negative workplace gossip → Turnover intentions (Model 4): [0.19, 0.35]
Anxiety → Turnover intentions (Model 4): [0.52, 0.72]

Note: Multicollinearity diagnostics indicated no serious concerns, with VIF values ranging from 1.08 to 2.31 and tolerance values ranging from 0.43 to 0.93 across the regression models.

Abbreviation: CI, confidence interval.

^** p < 0.01.

^*** p < 0.001.

To examine the mediating role of anxiety, negative workplace gossip and anxiety were entered simultaneously in Model 4. The results show that anxiety was significantly and positively related to turnover intentions (β = 0.62, p < 0.001). At the same time, the coefficient for negative workplace gossip decreased substantially, from β = 0.67 in Model 3 to β = 0.27 in Model 4, although it remained statistically significant (p < 0.001). These findings suggest that anxiety partially mediated the relationship between negative workplace gossip and turnover intentions, thereby supporting Hypothesis 3. In addition, multicollinearity diagnostics indicated no serious concerns, with all VIF values below the recommended threshold and all tolerance values within an acceptable range (see Table 5 note for details). The R ² values of 0.41 (for anxiety) and 0.68 (for turnover intentions) are moderate to high in nursing and behavioural research (Cohen 1988), indicating acceptable explanatory power of the model.

To further test the mediating role of anxiety, the SPSS PROCESS Macro Model 4 was utilised (Hayes 2013). The results of the bootstrap method test with 5000 bootstrap samples are presented in Table 6, indicating that the indirect effect of negative workplace gossip on turnover intentions through anxiety was significant (effect = 0.39) and the 95% bias‐corrected confidence interval (CI) of this effect was [0.31, 0.48], excluding 0, thereby providing further support for Hypothesis 3.

TABLE 6.

Test of the direct and indirect paths (N = 320).

Paths	Effect	SE	95% CI
			LLCI	ULCI
NWG → TOI	0.25	0.04	0.17	0.33
NWG → Anxiety → TOI	0.39	0.04	0.31	0.48

Abbreviations: CI, confidence interval; LLCI, lower limit confidence interval; NWG, negative workplace gossip; SE, standard error; TOI, turnover intentions; ULCI, upper limit confidence interval.

As a robustness check, we additionally tested the hypothesised mediation model using structural equation modelling (SEM) in AMOS 26.0. The results indicated an acceptable model fit (χ ² = 214.36, df = 73, χ ²/df = 2.94, CFI = 0.98, TLI = 0.97, RMSEA = 0.07). The SEM results were substantively consistent with the regression findings: negative workplace gossip was positively associated with anxiety (β = 0.64, p < 0.001), anxiety was positively associated with turnover intentions (β = 0.61, p < 0.001) and the direct path from negative workplace gossip to turnover intentions remained significant (β = 0.27, p < 0.001). The indirect effect through anxiety was also significant (indirect effect = 0.37, 95% CI [0.29, 0.46]), supporting the robustness of the partial mediation result.

4. Discussion

This study explored the impact of negative workplace gossip on turnover intentions among targeted nurses, with a particular focus on the mediating role of anxiety. Our findings indicate that negative workplace gossip was significantly associated with turnover intentions, both directly and indirectly through anxiety. The observed correlations among the study variables were relatively high (r = 0.64–0.79), which may reflect the inherent relatedness of workplace stressors and emotional outcomes in healthcare settings. However, the three‐factor CFA model demonstrated substantially better fit than alternative models (Table 2) and the HTMT values were below the conservative threshold of 0.85 (Table 3), supporting the discriminant validity of the constructs. These results align with existing literature on occupational stress and support theories of emotional contagion in workplaces, showing how adverse interpersonal interactions are associated with stress and dissatisfaction, ultimately impacting retention (Grande et al. 2024; Chang et al. 2019). This study contributes to understanding turnover mechanisms by emphasising that relational stressors, such as gossip, may be as impactful as structural job factors, underscoring the critical need to address workplace culture in healthcare settings.

Our study builds upon AET, which proposes that emotionally charged incidents in the workplace elicit strong affective reactions that shape employee attitudes and behaviours (Weiss and Cropanzano 1996). Negative gossip serves as a highly impactful affective event that is significantly related to sustained anxiety, which showed a positive association with turnover intentions. This dynamic is consistent with previous studies showing that interpersonal stressors are uniquely associated with reduced workplace well‐being as they may trigger emotional responses that are difficult to mitigate (Bakker and Demerouti 2017). However, unlike studies that focus on general organisational stressors such as workload or managerial support (Van Bogaert et al. 2014), our findings emphasise the relational dimension. Specifically, they show that hostile social interactions have a distinct and direct pathway to negative work outcomes, particularly in professions where teamwork and psychological safety are critical.

Moreover, the cultural context in which this study was conducted may amplify these effects. In collectivist cultures such as China, workplace relationships play a central role in shaping employees' sense of belonging, identity and psychological safety. Relational stressors like gossip carry stronger social consequences because they threaten group harmony and undermine individuals' standing within tightly interconnected workgroups. Prior research suggests that employees in collectivist settings are more sensitive to interpersonal evaluation and social exclusion, making negative gossip particularly detrimental to emotional well‐being and more likely to trigger anxiety‐driven withdrawal intentions. This cultural emphasis on social cohesion may therefore intensify the impact of gossip relative to more individualistic cultural contexts.

The mediating role of anxiety provides unique insights into the affective mechanisms linking workplace gossip and turnover intentions. Consistent with prior research, anxiety emerged as a potent intermediary statistically associated with the relationship between social stressors and behavioural responses, confirming its significance in withdrawal intentions within high‐stakes environments like healthcare (Horenstein and Heimberg 2020). In this context, anxiety reflects both the anticipated harm of social isolation and the perceived threat to professional reputation. These concerns were associated with a greater desire to leave as a way to preserve their psychological well‐being. This finding supports the broader occupational health literature, which suggests that while general workplace dissatisfaction can prompt turnover, targeted emotional responses like anxiety serve as a more immediate trigger for withdrawal behaviours, particularly in emotionally demanding roles (Delgado et al. 2017).

In addition, our findings extend research on turnover by showing that negative gossip is associated with a persistent emotional toll that standard workplace interventions may not adequately address. Prior studies have highlighted the benefits of resilience‐building programs in buffering stress from structural job demands (Hayes et al. 2012), yet such strategies may be less effective against the personalised and enduring nature of gossip‐related stress. Research on workplace conflict and relational aggression has similarly indicated that employees exposed to interpersonal hostility may face heightened psychological distress and isolation, ultimately undermining typical coping mechanisms and intensifying turnover intentions (Anasori et al. 2020). Our results align with this perspective, suggesting that healthcare organisations must consider specialised interventions—such as counselling, peer support and conflict resolution programs—to mitigate the unique challenges posed by negative gossip.

Furthermore, this study advances literature on workplace climate by emphasising the critical role of supportive social dynamics in retention. While traditional approaches to improving job satisfaction and retention have largely focused on structural improvements, such as reducing workloads or enhancing managerial support (Van Bogaert et al. 2014), our findings underscore the need for a positive relational climate. Specifically, negative gossip is negatively associated with team cohesion and psychological safety, both of which are essential for effective patient care. Studies have shown that fostering open communication and mutual trust can significantly mitigate the effects of interpersonal stressors, reducing turnover intentions and enhancing overall job satisfaction (Leger et al. 2022). By addressing relational stressors such as negative gossip, healthcare organisations may not only improve retention but also contribute to a more resilient and collaborative work environment that promotes high‐quality care.

Finally, the focus on anxiety as a mediator between gossip and turnover intentions offers valuable insights for healthcare administration. Previous studies on occupational stress have shown that anxiety may impair job performance and decision‐making by depleting cognitive and emotional resources (Colligan and Higgins 2006). Our study's findings suggest that, beyond job performance, anxiety stemming from workplace gossip is positively associated with withdrawal behaviours. This relationship between anxiety and turnover highlights the critical need for healthcare organisations to proactively address relational stressors that undermine psychological safety and foster a supportive culture. Interventions that promote team‐building, empathy training and open feedback mechanisms could play a significant role in mitigating these effects and supporting nurse retention.

These findings are particularly relevant to nursing management because negative gossip may undermine psychological safety, peer trust and communication within nursing teams. Nurse managers should therefore treat gossip not merely as an interpersonal issue but as a potential retention risk. Establishing clear norms for respectful communication, confidential reporting channels, timely conflict resolution and accessible psychological support may help reduce anxiety among targeted nurses and support workforce stability.

The findings from this study carry significant practical implications for healthcare administrators aiming to improve nurse retention and ensure a supportive work environment. High turnover rates among nurses not only result in substantial financial and operational burdens for healthcare organisations but also impact the continuity and quality of patient care. Therefore, it is crucial for healthcare organisations to recognise the role of negative interpersonal dynamics in influencing turnover intentions and take proactive steps to create a more positive work culture. Specifically, interventions targeting the reduction of negative workplace gossip and its psychological impact on employees may be essential in mitigating turnover.

To address negative workplace gossip, healthcare managers could implement policies that promote a culture of mutual respect and psychological safety, thereby reducing the likelihood of interpersonal hostility. Organisational interventions, such as team‐building activities, conflict resolution workshops and open communication channels, can foster an environment that discourages gossip and encourages positive social interactions. Additionally, training on emotional intelligence and resilience could help nurses manage anxiety effectively and improve their coping mechanisms when confronted with negative workplace interactions.

Further, the development of a strong support system within healthcare organisations is essential. Managers should provide mental health resources, such as counselling and stress management programs, to help nurses navigate the psychological challenges posed by workplace gossip. Regular feedback mechanisms that allow nurses to express their concerns and experiences anonymously may also help identify and address sources of interpersonal stress, thereby reducing the potential for gossip and its associated negative effects. Healthcare organisations could also consider incorporating mental health support directly into employee benefits packages to ensure that all staff members have access to resources for managing anxiety and related issues.

5. Limitations and Future Research Directions

While this study provides valuable insights, it is not without limitations. First, the cross‐sectional nature of this study limits our ability to infer causality between negative workplace gossip, anxiety and turnover intentions. Future research should consider employing longitudinal designs to explore how negative gossip and anxiety evolve over time and their long‐term effects on turnover intentions. A longitudinal approach would provide a more comprehensive view of the temporal relationship between these variables and allow researchers to observe changes in turnover intentions as workplace dynamics shift. Therefore, the mediation analysis should be interpreted as a theory‐driven statistical mediation test rather than evidence of causal or temporal mediation.

Second, this study relied on self‐report measures, which may be subject to social desirability bias. Participants might underreport their experiences with workplace gossip or anxiety due to concerns about judgement or reprisal. In addition, although the original GAD‐7 item content was retained, this study used a 5‐point response format to maintain consistency with the other measures. Future research should formally validate the psychometric properties of the 5‐point GAD‐7 format in Chinese nursing populations. Future research could also benefit from a multi‐source data approach, incorporating perspectives from peers, supervisors and organisational records to offer a more objective assessment of workplace interactions and turnover intentions. Additionally, future studies could consider the moderating effects of individual differences, such as resilience and coping mechanisms, to explore how personal factors influence the relationship between workplace gossip, anxiety and turnover intentions.

Lastly, this study was conducted within a specific cultural context, which may limit the generalisability of the findings to other healthcare settings. Cross‐cultural studies could provide valuable insights into how cultural norms and organisational structures shape the impact of workplace gossip and the associated psychological effects. Examining these relationships across different healthcare environments and cultural settings would enhance our understanding of how to mitigate negative interpersonal behaviours in diverse contexts.

6. Conclusion

In conclusion, this study provides evidence that negative workplace gossip is significantly associated with turnover intentions of targeted nurses, with anxiety acting as a mediating factor. These findings underscore the importance of addressing negative interpersonal behaviours within healthcare organisations to improve nurse retention and overall workplace well‐being. To mitigate the impact of negative gossip, healthcare managers may consider implementing specific interventions such as establishing clear anti‐gossip policies, offering conflict‐resolution and communication training, providing confidential counselling and psychological support services and creating structured peer‐support programs that strengthen team cohesion. By fostering a supportive and respectful work culture, reducing sources of interpersonal stress and providing robust mental health resources, healthcare organisations can work towards enhancing nurse satisfaction and retention.

Author Contributions

Ying Li: conceptualisation, methodology, writing – review and editing. Tao Shen: supervision, conceptualisation, methodology, writing – original draft. Xiaolin Shen: data curation, formal analysis, investigation, writing – review and editing. All authors have read and approved the final manuscript.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not‐for‐profit sectors.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: Supporting Information.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.