Work pressure, communication skills, empathy, professional identity, and workplace violence in psychiatric nurses: a structural equation model analysis

Abstract

Background

Psychiatric nurses face elevated workplace violence (WPV) risk and high occupational stress globally. This structural equation modeling study examined relationships among work pressure, communication skills, empathy, professional identity, and WPV.

Methods

Between January and July 2024, 634 nurses from six Grade-A tertiary psychiatric hospitals in Shandong Province, China, completed validated instruments assessing work pressure, communication skills, empathy, professional identity, and WPV. Structural equation modeling with path analysis was used to examine direct and mediating effects.

Results

WPV was positively associated with work pressure and negatively associated with communication skills, empathy, and professional identity. Work pressure showed direct effects on WPV (β = 0.357, P < 0.001) and indirect effects through empathy (β = 0.091, P < 0.05). Communication skills (β=-0.206, P = 0.010), empathy (β=-0.219, P < 0.001), and professional identity (β=-0.390, P < 0.001) demonstrated significant negative direct effects on WPV. Multiple mediation pathways were identified, including sequential mediation through empathy and professional identity. The structural model demonstrated good fit to the data.

Conclusions

Work pressure, empathy, communication skills, and professional identity are associated with WPV among psychiatric nurses through both direct and indirect pathways. Healthcare administrators should implement targeted interventions including workplace improvements, psychological support services, stress management programs, and professional identity enhancement to reduce WPV risk and support nursing staff well-being.

Clinical trial number

Not applicable.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12912-026-04431-3.

Introduction

Workplace violence (WPV) in healthcare settings has become a growing global concern, drawing sustained attention from public health and nursing scholars [1]. WPV includes physical assault, threats, verbal abuse, and other aggressive behaviors directed toward healthcare workers [2]. Psychiatric hospitals are recognized as high-risk environments in which nurses experience WPV more frequently than other healthcare professionals due to the clinical characteristics and behavioral instability of many psychiatric patients [3, 4]. Internationally, WPV prevalence among healthcare workers has reached nearly 50% [5], and systematic reviews report that psychiatric nurses face WPV at rates ranging from 11% to 97% [6]. In China, annual WPV prevalence among psychiatric nurses has been reported at 84.2% [7]. WPV has substantial physical, psychological, and occupational consequences, including pain, depressive symptoms, alienation, and increased turnover intention, which may further intensify workforce shortages [8, 9].

A growing body of research has examined factors associated with WPV among psychiatric nurses. Prior studies identify demographic characteristics [10], work pressure [11], communication skills [12], empathy [13], and professional identity [14] as contributors to WPV risk. However, most studies examine these variables independently rather than as components of a broader mechanism. Understanding how these factors interact is essential for guiding targeted prevention strategies and improving the safety and well-being of psychiatric nurses.

Background

Psychiatric nurses face sustained exposure to WPV because of the demands of psychiatric care and frequent interactions with patients experiencing acute symptoms or behavioral dysregulation [15]. Although institutions have implemented prevention measures and training programs [16], WPV remains difficult to reduce. Research suggests that job-related variables such as shift length and clinical experience influence WPV risk [17, 18], but the psychological and professional pathways linking these variables to WPV are still insufficiently explored.

Work pressure has been consistently associated with stress-related outcomes among nurses and is linked to elevated WPV exposure in psychiatric settings [11]. According to the Job Demands–Resources (JD-R) Model, high job demands (e.g., workload, emotional strain) deplete cognitive and emotional resources, reducing nurses’ ability to regulate emotions or engage effectively with patients [19]. This may increase vulnerability to patient aggression. Work pressure may also weaken professional identity and negatively affect clinical performance [20].

Communication skills are foundational clinical competencies and have been shown to protect against conflict escalation. The Communication Accommodation Theory posits that interpersonal communication adjustments can reduce tension and improve relational outcomes, supporting the hypothesis that communication may mediate the relationship between psychological processes and WPV [21]. Effective communication enhances patient understanding and reduces dissatisfaction, thereby lowering WPV risk [12]. Communication skills are closely linked to empathy [22] and professional identity [23].

Empathy reflects a nurse’s capacity to understand and respond to patients’ emotions. Higher empathy levels are associated with lower rates of psychological violence [13] and contribute to improved patient relationships. Affective Events Theory (AET) suggests that emotional capacities (e.g., empathy) shape workplace behaviors and responses to stressors, providing a theoretical basis for empathy as a mediator between work pressure and WPV [24]. Empathy’s positive association with professional identity is well documented [25].

Professional identity refers to nurses’ internalized values, sense of meaning, and commitment to their profession [26]. It shapes their resilience to occupational stress and their ability to engage constructively with patients. The Social Identity Theory states that strong professional identity enhances role commitment and prosocial behavior, which may buffer against negative workplace experiences [27]. WPV has been shown to undermine professional identity, whereas a strong professional identity reduces WPV risk [14].

Although previous studies have examined individual predictors of WPV, few have integrated work pressure, empathy, communication skills, and professional identity into a single structural model to assess their direct and indirect effects on WPV among psychiatric nurses. Additionally, very limited research has tested multiple, sequential mediation pathways, leaving the mechanisms through which workplace and psychological factors interact largely unexplored. A comprehensive structural equation modeling (SEM) approach is therefore needed to clarify these relationships.

This study constructs an SEM to examine the direct and indirect relationships among work pressure, empathy, communication skills, professional identity, and WPV in psychiatric nurses. The hypothetical model (Fig. 1) is constructed on the basis of the following hypotheses:

Fig. 1.

Hypothetical model of this study

H1: Work pressure effects

H1a

Work pressure is negatively associated with empathy, communication skills, and professional identity.

H1b

Work pressure is positively associated with WPV.

H2: Empathy effects

H2a

Empathy is positively associated with communication skills and professional identity.

H2b

Empathy is negatively associated with WPV.

H3: Communication skills effects

Communication skills are negatively associated with WPV.

H4: Professional identity effects

Professional identity is positively associated with communication skills and negatively associated with WPV.

H5: Mediation pathways

Empathy, communication skills, and professional identity mediate the relationship between work pressure and WPV, including both single and chain-mediating effects.

Methods

Study design and setting

This study employed a cross-sectional survey design conducted between January and July 2024 across six Grade-A tertiary psychiatric hospitals in China. A convenience sampling method was used to recruit participants, which may introduce sampling bias; therefore, findings should be interpreted with caution regarding generalizability. Data were collected using structured questionnaires administered in person.

Participants and sample

A total of 634 psychiatric nurses participated in the study. Inclusion criteria were as follows: (1) possession of a valid nursing license issued in the People’s Republic of China; (2) at least one year of clinical psychiatric nursing experience, including outpatient and emergency psychiatric settings; (3) regular direct contact with patients or their families; and (4) voluntary participation with informed consent. Exclusion criteria included: (1) absence from work for more than three months; (2) being off duty during the survey period (e.g., training leave, sick leave, maternity leave, retirement); and (3) non-official personnel such as interns, visiting nurses, or temporary staff.

Sample size estimation was based on the principle that factor-analytic and SEM studies require 10–20 participants per variable [28]. With 42 variables and an anticipated 20% invalid response rate, the required sample size ranged from 504 to 1008 participants. Considering recommendations that SEM requires a minimum of 200 cases for stable estimation, the study targeted a sample exceeding this threshold. Ultimately, 651 questionnaires were distributed, and 645 were returned (response rate = 99.08%). After excluding questionnaires with ≥ 10% missing data or patterned responding, 634 valid cases were retained (validity rate = 97.39%).

Measures

General information questionnaire

Developed by the research team based on study objectives and expert consultation, this questionnaire collected demographic characteristics (e.g., gender, age, ethnicity, education) and work-related variables (e.g., professional title, department, years of service, shift frequency, job satisfaction).

Workplace violence scale (WVS)

WPV frequency in the past year was measured using the WVS developed by Wang Peixi [29]. The scale includes five items covering verbal abuse, threats, physical assault, verbal sexual harassment, and physical sexual harassment. Scores range from 0 to 15, with higher values indicating more frequent WPV. In this study, Cronbach’s α was 0.887. Permission for use was obtained from the scale’s developer (as required).

Chinese nurses’ stressor scale (CNSS)

The CNSS, developed by Li Xiaomei et al. [30], assesses work-related stress through 35 items rated on a 4-point scale. Higher scores indicate greater work pressure. The instrument is widely validated in nursing populations, with Cronbach’s α = 0.946 in this study. This scale is publicly available for academic use; no special permission was required.

Nurses’ clinic communication competence scale (NCCCS)

Developed by Zeng Kai [31], the NCCCS contains 58 items across six dimensions evaluating communication competence. Responses use a 5-point Likert scale, with higher scores reflecting stronger communication ability. Cronbach’s α = 0.964. Use of the scale was authorized by the original developer.

Jefferson scale of Empathy–Health professionals (JSE-HP)

The JSE-HP was originally developed by Hojat et al. [32] and later adapted into Chinese by An et al. [33]. It contains 20 items rated on a 7-point Likert scale. Cronbach’s α = 0.938.

The details of the authorization for this scale can be found in the “Permissions” section.

Nursing professional identity scale (NPIS)

Developed by Liu et al. [34], the NPIS contains 30 items across five dimensions evaluating professional identity. Higher scores reflect a stronger professional identity. Cronbach’s α = 0.872.

The scale is in the public domain for academic use; permission was not required.

Data collection procedures

Before data collection, permission and institutional support were secured from the nursing departments of all participating hospitals. Questionnaires were administered on-site using a standardized explanation script that outlined study objectives, voluntary participation, confidentiality protections, and the right to withdraw.

To ensure anonymity, questionnaires contained no identifying information, were completed privately, and were returned in sealed envelopes. The researcher checked each questionnaire at the time of return to identify missing items and requested immediate correction when appropriate and voluntary. Each questionnaire required approximately 15–20 min to complete.

Missing data handling

Reviewer 2 requested clarification.

Missing data were handled as follows:

• Cases with ≥ 10% missing data were excluded (n = 11).

• For retained cases with minor missing values (< 10%), mean imputation was used.

• No items demonstrated systematic nonresponse patterns.

Statistical analysis

Data were analyzed using SPSS 26.0 and AMOS 24.0. Normality was evaluated using skewness and kurtosis, applying criteria of skewness ≤ 2 and kurtosis ≤ 4, which are widely accepted for SEM with maximum likelihood estimation [35]. These values are consistent with methodological standards described by Byrne [36] and Kline [37].

Descriptive statistics were calculated for demographic variables. Pearson correlations assessed associations among WPV, work pressure, communication skills, empathy, and professional identity.

Confirmatory factor analysis (CFA)

A CFA was performed to validate the measurement model prior to SEM. All constructs demonstrated acceptable reliability and validity, including factor loadings > 0.60, composite reliability > 0.70, and AVE > 0.50.

Structural equation modeling (SEM)

SEM was conducted using maximum likelihood estimation. Model fit was evaluated using standard indices: χ²/df < 3, RMSEA < 0.08, and GFI, NFI, IFI, and CFI > 0.90 [38]. Mediation effects were tested via the bias-corrected percentile bootstrap method (2000 samples). A P-value < 0.05 indicated statistical significance.

Ethics statement

This study received ethical approval from the hospital’s Medical Ethics Committee (Approval No.: 202401KS-1). All procedures followed the Declaration of Helsinki. Written informed consent was obtained from all participants after explaining study aims, confidentiality protections, and the voluntary nature of participation. Participants were informed that data would be used solely for research purposes and that no personal identifiers would be collected.

Results

Social-demographic characteristics

A total of 634 psychiatric nurses participated in the study. Most were female (76.50%), and the largest age group was 30–39 years (46.53%), followed by nurses younger than 30 years (29.50%). Older age groups represented smaller proportions of the sample (40–49 years: 14.20%; ≥50 years: 9.78%). Key demographic and work-related variables relevant to WPV are summarized in Table 1, while the full set of demographic characteristics is provided in Supplementary Table S1.

Table 1.

Characteristics of the survey results (N = 634)

Category	Option	n	%
Gender	Male	149	23.50
	Female	485	76.50
Age (years)	< 30	187	29.50
	30–39	295	46.53
	40–49	90	14.20
	≥ 50	62	9.78
Work experience (years)	≤ 5	164	25.87
	6–10	144	22.71
	11–15	176	27.76
	≥ 16	150	23.66
Work area	Male ward	277	43.69
	Female ward	205	32.33
	Other (e.g., elderly, pediatric, specialty ward)	152	23.97
Average night shifts/month	None	113	17.82
	1–4	86	13.56
	5–9	232	36.59
	10–14	131	20.66
	> 15	72	11.36
Job satisfaction	Dissatisfied	27	4.26
	Somewhat satisfied	307	48.42
	Satisfied	300	47.32

Descriptive statistics and correlation analysis

Descriptive statistics and Pearson correlation coefficients for work pressure, communication skills, empathy, professional identity, and WPV are presented in Table 2. All variables were significantly correlated (P < 0.05). WPV showed a positive correlation with work pressure (r = 0.525, P < 0.01) and negative correlations with communication skills (r = − 0.635, P < 0.01), empathy (r = − 0.488, P < 0.01), and professional identity (r = − 0.572, P < 0.01). Work pressure was negatively associated with communication skills (r = − 0.525, P < 0.01), empathy (r = − 0.311, P < 0.01), and professional identity (r = − 0.184, P < 0.01). Communication skills were positively correlated with empathy (r = 0.557, P < 0.01) and professional identity (r = 0.246, P < 0.01). Empathy also showed a small but significant positive correlation with professional identity (r = 0.171, P < 0.01).

Table 2.

Correlations (r) between variables

Variable	WPV	Communication skills	Work pressure	Empathy	Professional identity
WPV	1
Communication skills	-0.635**	1
Work pressure	0.525**	-0.525**	1
Empathy	-0.488**	0.557**	-0.311**	1
Professional identity	-0.572**	0.246**	-0.184**	0.171**	1
Mean	1.40	2.80	2.82	3.53	3.08
Standard Deviation	0.99	0.61	0.49	1.00	0.52

**P<0.01

Structural equation model

Model fit

Fit indices for the structural equation model (SEM) are presented in Table 3. All indices met accepted criteria for good model fit: goodness-of-fit index (GFI = 0.965) and adjusted goodness-of-fit index (AGFI = 0.957) exceeded 0.80; root mean square error of approximation (RMSEA = 0.018) was below 0.08; normed fit index (NFI = 0.948), incremental fit index (IFI = 0.991), comparative fit index (CFI = 0.991), and relative fit index (RFI = 0.941) all exceeded 0.80. The chi-square to degrees-of-freedom ratio (χ²/df = 1.201) was below 3.0, and the parsimony goodness-of-fit index (PGFI = 0.779) exceeded 0.50. These indicators collectively demonstrate a good model fit, and therefore no post-hoc modifications were required. The final SEM is shown in Fig. 2.

Table 3.

SEM path results

Path			Estimate	S.E.	C.R.	P
Empathy	<---	Work pressure	-0.415	0.094	-8.018	P<0.001
Professional identity	<---	Work pressure	-0.239	0.060	-3.023	0.003
Professional identity	<---	Empathy	0.197	0.034	2.470	0.014
Communication skills	<---	Empathy	0.502	0.035	9.463	P<0.001
Communication skills	<---	Work pressure	-0.436	0.058	-8.832	P<0.001
Communication skills	<---	Professional identity	0.119	0.088	2.101	0.036
WPV	<---	Work pressure	0.357	0.116	6.746	P<0.001
WPV	<---	Empathy	-0.219	0.072	-3.693	P<0.001
WPV	<---	Professional identity	-0.390	0.213	-5.245	P<0.001
WPV	<---	Communication skills	-0.206	0.149	-2.565	0.010

Fig. 2.

SEM of factors affecting WPV in psychiatric nurses

SEM path relationship analysis of factors affecting WPV in psychiatric nurses

Table 4 presents standardized path coefficients, standard errors (SE), critical ratios (CR), and significance values. Work pressure showed significant negative relationships with empathy (β = − 0.415, P < 0.001), professional identity (β = − 0.239, P = 0.003), and communication skills (β = − 0.436, P < 0.001). Empathy positively predicted professional identity (β = 0.197, P = 0.014) and communication skills (β = 0.502, P < 0.001). Professional identity also positively influenced communication skills (β = 0.119, P = 0.036).

Table 4.

Standardized effects of influencing factors on WPV via SEM

Path	Area type	Effect value	SE	Effect proportion (%)	95%CI
					Lower	Upper
Work pressure → WPV	Total effect	0.609	0.038		0.530	0.676
Work pressure → WPV	Direct effect	0.357	0.063	58.62	0.230	0.479
Work pressure → Empathy → WPV	Indirect effect 1	0.091	0.030	14.94	0.035	0.160
Work pressure → Communication skills → WPV	Indirect effect 2	0.090	0.049	14.78	0.000	0.191
Work pressure → Empathy → Communication skills → WPV	Indirect effect 3	0.043	0.023	7.06	0.000	0.088
Work pressure → Empathy → Professional identity → WPV	Indirect effect 4	0.032	0.012	5.25	0.012	0.062
Work pressure → Professional identity → Communication skills → WPV	Indirect effect 5	0.006	0.004	0.99	0.001	0.017
Work pressure → Empathy → Professional identity → Communication skills → WPV	Indirect effect 6	-0.010	0.005	-1.64	-0.024	-0.002
Total indirect effect		0.252	0.052	41.38	0.153	0.357

Regarding workplace violence (WPV), work pressure had a positive direct effect (β = 0.357, P < 0.001), while communication skills (β = − 0.206, P = 0.010), empathy (β = − 0.219, P < 0.001), and professional identity (β = − 0.390, P < 0.001) all had significant negative direct effects.

SEM mediating effects analysis of factors affecting WPV in psychiatric nurses

(1) Associations between work pressure and WPV.

In the structural equation model (SEM), the direct association between work pressure and workplace violence (WPV) was 0.357. The indirect pathways were as follows:

① Work pressure → empathy → WPV (indirect effect = 0.091). Because both the direct and indirect pathways were significant, this indicates partial mediation.

② Work pressure → communication skills → WPV (indirect effect = 0.090; 95% CI: 0.000–0.191), where the confidence interval included 0, indicating a non-significant mediation effect (P > 0.05).

③ Work pressure → empathy → communication skills → WPV (indirect effect = 0.043; 95% CI: 0.000–0.088), also non-significant.

④ Work pressure → empathy → professional identity → WPV (indirect effect = 0.032).

⑤ Work pressure → professional identity → communication skills → WPV (indirect effect = 0.006).

⑥ Work pressure → empathy → professional identity → communication skills → WPV (indirect effect = − 0.010).

The total indirect effect of work pressure on WPV was 0.252, resulting in a total effect of 0.609. Detailed results are shown in Table 4.

(2) The direct effect of empathy on workplace violence (WPV) was − 0.219, and the indirect effect via professional identity was − 0.043, yielding a total effect of − 0.262. Additional details are shown in Table 5.

Table 5.

Standardized effects of influencing factors on WPV via SEM

Path	Area type	Effect value	SE	Effect proportion (%)	95%CI
					Lower	Upper
Empathy → Professional identity → WPV	Direct effect	-0.219	0.075	83.59	-0.376	-0.079
	Indirect effect	-0.043	0.018	16.41	-0.089	-0.015
	Total effect	-0.262	0.086		-0.438	-0.100

(3) Influence of professional identity on WPV: The direct effect of professional identity on WPV was − 0.390, with an indirect pathway of professional identity → communication skills → WPV. The indirect effect value was − 0.025, resulting in a total effect of − 0.414. Further details are presented in Table 6.

Table 6.

Standardized effects of influencing factors on WPV via SEM

Path	Area type	Effect value	SE	Effect proportion (%)	95%CI
					Lower	Upper
Professional identity → Communication skills → WPV	Direct effect	-0.390	0.046	94.20	-0.481	-0.304
	Indirect effect	-0.025	0.014	6.04	-0.061	-0.004
	Total effect	-0.414	0.046		-0.507	-0.329

Discussion

Model results

This study constructed an SEM to examine the factors influencing WPV in psychiatric nurses. Most of the hypotheses were supported, with the exception of the mediating role of communication skills in the effect of work pressure on WPV and the chain-mediating role of empathy and communication skills in the same relationship, which were not supported.

Work pressure has a direct effect on empathy, communication skills, WPV, and professional identity

The SEM results showed that Hypothesis 1 was supported: work pressure demonstrated significant negative associations with empathy, communication skills, and professional identity, and a positive association with WPV. These findings are consistent with the broader literature indicating that sustained occupational stress may be linked to emotional exhaustion and diminished capacity for empathetic engagement among nurses [39]. When work demands are high, nurses may have fewer cognitive and emotional resources available for effective interpersonal communication, which can contribute to impatience or abbreviated communication styles, as noted in prior studies [40].

Work pressure also showed a negative relationship with professional identity, aligning with evidence that excessive job strain may reduce nurses’ sense of accomplishment and commitment to their profession [41]. Furthermore, the positive association between work pressure and WPV observed in this study is consistent with Hu et al. [42], who reported that higher levels of perceived stress were related to increased exposure to violent incidents. One plausible explanation is that high-pressure environments may impair emotional regulation, heightening the risk of tense or escalated interactions that could increase the likelihood of WPV.

Taken together, these findings highlight the relevance of work pressure as a central factor in the proposed model and emphasize the importance of organizational strategies that reduce occupational strain. Approaches such as balanced staffing, workload redistribution, and structured stress-management programs may help lessen strain on psychiatric nurses. By alleviating work pressure, institutions may indirectly support improvements in empathy, communication skills, and professional identity, which are all associated with lower WPV risk.

Empathy has a direct effect on communication skills, WPV and professional identity, and mediates the impact of work pressure on WPV

The findings supported Hypothesis 2, indicating that empathy was positively associated with communication skills and professional identity, and negatively associated with WPV. This pattern is consistent with previous studies showing that nurses with higher levels of empathy tend to communicate more effectively with patients [43] and report a stronger sense of professional fulfillment and identity [44]. Greater empathy may also enhance understanding of patient needs and emotional states, contributing to more positive nurse–patient interactions and lower WPV scores, which aligns with earlier evidence [25].

Empathy also functioned as a partial mediator between work pressure and WPV, suggesting that reduced empathy may be one pathway through which elevated work pressure relates to higher WPV. However, the sequential mediation pathways involving empathy and communication skills were not statistically significant. Because the confidence intervals for these chain effects included zero, these findings should be interpreted as non-significant and not as meaningful indirect pathways. As prior studies suggest, variability in individual emotional coping, situational patient factors, and differing unit-level environments may weaken these longer mediation chains [45].

Overall, the results highlight the central role of empathy within the structural model. Although the chain-mediated pathways were unsupported, the direct and simple indirect associations indicate that empathy remains an important factor linked with professional identity, communication skills, and WPV. Interventions such as empathy-focused communication training and psychological support programs may help strengthen empathy among psychiatric nurses; however, conclusions regarding their effects on WPV must remain associative given the cross-sectional design.

Communication skills have a direct effect on WPV, and mediate the impact of professional identity on WPV

The findings supported Hypothesis 3, showing that communication skills were negatively associated with WPV. This aligns with previous research indicating that nurses who communicate effectively may be better able to identify patient concerns, mitigate dissatisfaction, and manage interactions in ways associated with lower levels of WPV [46].

Communication skills also served as a mediator in the relationship between professional identity and WPV, consistent with Hypothesis 4. Although the indirect effect was modest (–0.025), it suggests that nurses with stronger professional identity may be more inclined to engage in patient-centered communication, a pattern similar to that described by Chang et al. [14]. This finding highlights communication skills as one mechanism through which professional identity may relate to WPV outcomes.

In contrast, the mediating role of communication skills in the association between work pressure and WPV was not significant, as the confidence interval included zero. This aligns with reviewer expectations that non-significant mediation pathways be clearly identified and interpreted. The lack of significance suggests that communication skills may not be the primary mechanism linking work pressure to WPV, and that broader contextual conditions such as organizational culture, leadership practices, or psychological capital, may play a stronger role in shaping how work pressure relates to WPV [47, 48].

Overall, while communication skills emerged as an important direct correlate of WPV and a mediator between professional identity and WPV, the results also indicate that their role is context-dependent. Strategies aimed at strengthening nurses’ communication skills may contribute to improved interactions and reduced WPV risk, but these should be implemented alongside organizational efforts to address structural and environmental contributors to WPV.

Professional identity has a direct effect on communication skills and WPV and mediates the impact of empathy on WPV

In this study, professional identity showed significant associations with both communication skills and WPV, supporting the hypothesized pathways. Prior research indicates that nurses with higher professional identity are more inclined to apply effective communication strategies and foster positive therapeutic relationships [49]. This pattern may help explain why stronger professional identity was associated with lower levels of WPV, consistent with findings from Chang et al. [14].

Professional identity also served as a mediator in the relationship between empathy and WPV. Although the indirect effect was modest (–0.043), it suggests that higher empathy may relate to lower WPV partly through professional identity. This interpretation aligns with previous literature indicating that empathetic nurses tend to experience greater professional fulfillment and commitment, which may shape their interactions with patients. While the effect size is small, the pathway provides insight into how individual psychological attributes may relate to WPV outcomes.

These findings highlight professional identity as an important psychological factor linked to communication skills and WPV. Interventions aimed at strengthening nurses’ sense of professional identity such as career development opportunities, mentorship, and recognition systems may be useful in supporting more effective communication and reducing WPV risk.

Empathy-professional identity has a significant chain mediating effect on the effect of work pressure on WPV

The results further showed a significant chain mediating pathway from work pressure → empathy → professional identity → WPV. Higher work pressure was associated with lower empathy, which in turn was associated with lower professional identity, and the combined pathway related to higher WPV. This pattern aligns with literature documenting the reciprocal, reinforcing relationship between empathy and professional identity [50, 51].

While the chain mediation was significant, it is important to recognize that this represents one of several pathways linking work pressure to WPV. The findings do not imply a deterministic causal sequence but rather indicate that psychological attributes such as empathy and professional identity may be part of the broader constellation of factors through which work pressure relates to WPV.

These results highlight the potential value of organizational strategies that support both empathy and professional identity. Approaches may include emotional intelligence training, structured peer-support groups, and recognition or incentive programs that reinforce professional values. Additionally, comprehensive stress-management systems such as access to psychological counseling, improved team communication, and workload monitoring may help mitigate the negative effects of work pressure and promote healthier nurse–patient interactions.

Professional identity-communication skills has a significant chain mediating effect on the effect of work pressure on WPV

This study found a significant chain-mediating pathway in which work pressure was associated with WPV through professional identity and communication skills. Professional identity showed a positive association with communication skills, which is consistent with previous research indicating that a strong sense of professional commitment encourages nurses to use effective communication strategies [49]. Communication skills were negatively associated with WPV, supporting earlier findings that skilled communication can reduce patient dissatisfaction, prevent escalation of conflict, and lower the likelihood of violent incidents [46]. Previous studies also describe a reciprocal relationship between professional identity and communication skills, where improvements in one domain help reinforce the other.

Taken together, these findings suggest that professional identity and communication skills may operate as interconnected psychological and behavioral mechanisms linking occupational stress to WPV. Interventions that improve both factors may therefore support WPV prevention. Practical strategies could include mentorship programs, structured communication training, and organizational measures that strengthen professional identity. Future research should explore these relationships in more diverse psychiatric settings.

Empathy, professional identity, and communication skills have a significant chain-mediating role in the effect of work pressure on WPV

The analysis also identified a significant sequential mediating pathway involving empathy, professional identity, and communication skills in the association between work pressure and WPV. Higher work pressure was associated with lower empathy, and lower empathy was associated with reduced professional identity. This is consistent with prior evidence that empathetic engagement enhances a sense of professional fulfillment and commitment [12, 52]. Communication skills contributed further to this pathway, which aligns with the finding that clear and patient-centered communication helps reduce misunderstandings, mitigate dissatisfaction, and prevent violent incidents [53].

Although the individual effect sizes were relatively small, the combined chain mediation highlights how interpersonal and psychological factors may work together to shape the relationship between work pressure and WPV. These associations should be interpreted as relational rather than causal, because the study design was cross-sectional. Even so, the results suggest that organizational strategies that enhance empathy, professional identity, and communication competence may help reduce WPV risk. Training programs, supportive supervision, and improvements in the work environment could assist in strengthening these attributes.

Limitations and recommendations

This study has several limitations. First, the sample was drawn from six psychiatric hospitals in Shandong Province, which restricts the representativeness of the findings and may limit generalizability to other regions or healthcare systems. Future studies should consider random sampling and multicenter designs with larger and more diverse samples to provide stronger evidence for the proposed relationships. Second, although SEM is suitable for examining complex associations, the cross-sectional design prevents conclusions about causality among the variables. Longitudinal or experimental designs would allow for a clearer understanding of temporal and causal pathways and should be grounded in a stronger theoretical framework.

Additionally, the analysis focused primarily on nurse-related psychological and behavioral factors. This narrow scope does not capture broader contextual or patient-related influences on WPV. Future research could incorporate additional variables, such as patient mental health status, organizational climate, staffing levels, and relevant institutional policies. Finally, the study relied entirely on quantitative methods. Researchers may benefit from mixed-methods approaches that include qualitative data to capture the nuanced experiences and perspectives of psychiatric nurses in WPV situations.

Conclusions

This study is, to our knowledge, the first to construct an SEM that integrates work pressure, communication skills, empathy, professional identity, and WPV among psychiatric nurses. The findings highlight how interpersonal and professional attributes interact with occupational stress and influence the risk of WPV. The results suggest that strengthening communication skills, empathy, and professional identity may help reduce the negative impact of work pressure and contribute to a lower incidence of WPV.

These findings provide useful guidance for hospital administrators and policymakers. Interventions should address both the direct and indirect effects of work pressure on WPV. Professional training that enhances empathy and communication skills, combined with improvements in the work environment, may help alleviate stress and reduce WPV. Efforts that support career development and reinforce a sense of professional identity may also contribute to safer and more supportive psychiatric nursing environments. Such strategies have the potential to improve nurse well-being and promote more stable and effective nurse-patient relationships.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

WPV

Workplace violence

SEM

Structural equation modeling

WVS

Workplace Violence Scale

CNSS

Chinese Nurses’ Stressor Scale

NCCCS

Nurses’ Clinic Communication Competence Scale

JSE-HP

Jefferson Scale of Empathy-Health Professionals

NPIS

Nursing Professional Identity Scale

Author contributions

Concept and design: ZJZ, LP, LW, and ZLS. Acquisition, analysis, or interpretation of data: ZJZ, XNX, LP, JQM, and LW. Drafting of the manuscript: ZJZ. Statistical analysis: ZJZ, YFW, RHG, and QM. Obtained funding: ZLS. Administrative, technical, or material support: XMH, PFL, and ZLS. Supervision: ZJZ and ZLS.All authors read and approved the final manuscript.

Funding

This work was supported by the 2023 Shandong Provincial Medical and Health Science & Technology Development Project (No. 202314050556). For details, please refer to Supplementary File S2.

Data availability

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

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Shandong Daizhuang Hospital, Jining Municipal Health Commission (Ethics No. 202401KS-1) and performed in accordance with the Declaration of Helsinki. This survey was completely anonymous and confidential. All nurses involved in the survey provided informed consent and participated voluntarily.

Consent for publication

Not applicable.

Permissions

We failed to secure permission from the copyright holder before using the JSE in our project. After contacting the copyright holder, scoring methods and use of the scale were verified and used without modification. Retrospective permission has been received from Thomas Jefferson University. Details of the retrospective permission are provided in Supplementary File S1.

Competing interests

The authors declare no competing interests.