Background
Sleep is a physiological need of the body to sustain life and is essential for biological rhythms [1]. Poor sleep quality increases an individual’s risk of developing chronic diseases such as hypertension and diabetes, affects cognitive and daytime performance, and can lead to mental health problems such as depression and anxiety [2]. Therefore, good sleep quality is important for personal, professional, social, and family life [3].
In addition to factors such as gender, age, sleep environment, and physical characteristics, sleep quality is influenced by stressful events and psychological states [4]. Psychiatric nurses frequently encounter patients with mental health disorders. Consequently, psychiatric nurses experience prolonged periods of high-stress intensity and workload, along with frequent incidents of workplace violence [5], which can lead to negative emotions, increased stress levels, and a higher incidence of poor sleep quality [6]. Psychiatric nurses exhibit significantly higher rates of poor sleep than the general population, doctors, and other hospital staff [7, 8].
Anzeige
Psychiatric nurses are approximately 1.47 times more likely to experience workplace violence than physicians [9]. Violence at the workplace may lead to decreased sleep quality for nurses [10]. A meta-analysis revealed that workers exposed to violence exhibited an elevated risk of sleep disturbances (OR = 2.55) [11]. Workplace violence causes emotional responses that further influence attitudes and behaviours [12]. Internal experiences, such as negative emotions, can be expressed externally through facial expressions, posture, and language, which, in turn, can lead nurses to behave in ways that are incompatible with the work environment, such as arguing and being impatient [13]. As a profession involving close contact with patients and their families, nursing has fundamental professional morals. Hospital management requires nurses to express their emotions at work according to proper codes of conduct, such as smiling and showing empathy; this usually involves emotional labour [14].
Emotional labour is defined as a form of work in which individuals use affect and emotions to achieve organisational goals; in a work context, this means that employees are required to display emotional states at work that are satisfactory to the organisation [15]. Simultaneously, workplace violence can trigger negative emotions such as fear and anger [16], which usually require nurses to perform some amount of emotional labour to meet organisational needs. Therefore, workplace violence can be considered a negative workplace event [17]. Research has found significantly higher levels of emotional labour when employees are exposed to workplace violence [18]. Bai et al. [16] have demonstrated the mediating role of emotional labour in the relationship between patient incivility and nurse fatigue. Therefore, we hypothesise that surface and deep acting in emotional labour mediates the relationship between workplace violence and poor sleep quality among nurses.
Gross [19] argues that there are two stages of emotion control: (1) Simply controlling the outward expression of emotion, called expression suppression; and (2) Cognitively altering an individual’s understanding of an emotional event, called cognitive reappraisal. Surface acting suppresses expression in emotion regulation, whereas deep acting promotes cognitive reappraisal [15]. There are differences in the resources required to engage in different types of emotional labour. Expressing restraint involves continuous self-cueing and processing, such as telling oneself ‘I need to stay calm’, and is therefore more cognitively taxing than deep acting [19].
Zapf [20] has applied the action theory to the emotional labour theory and posited that surface acting occurs more often in habitual, routine processes, which require minimal conscious processing and may be completed semi-automatically. By contrast, deep acting regulates inner feelings and necessitates conscious processing, consuming more internal energy. For nurses, surface acting is more comfortable and easier because it does not require as much effort to regulate the mind as deep acting. People choose surface acting when their external emotions conflict with their internal experiences, leading to feelings of dissociation and emotional dissonance [21]. Deep acting prevents individuals from feeling emotionally dysregulated and reduces the stress and strain of emotional labour. Thus, people who choose deep instead of surface acting for emotional labour in the presence of negative emotions have a more positive impact regarding short-term affective, cognitive, and social functioning and well-being [22] and consume fewer resources. Consequently, we hypothesise that workplace violence results in different types of emotional work, which, in turn, has different effects on sleep quality.
Anzeige
Occupational identity refers to a practitioner’s self-concept based on attributes, beliefs, values, motivations, and experiences [23]. It is a positive resource that improves emotion regulation [24] and alleviates negative emotions at work [25]. According to the social identity theory, an individual’s emotional expressions in the workplace are affected by their level of occupational identity [26]. Individuals who strongly identify with their occupation tend to exhibit emotions aligned with their professional role, serving as both a symbol and safeguard of their occupational identity. However, following a negative experience, these employees may feel their development is at risk and reduce their efforts owing to risk aversion tendencies [27]. Their level of occupational identity may influence the choice of emotion regulation strategies among nurses when they encounter negative stimuli such as workplace violence.
The relationships between violence, emotional labour, occupational identity, and sleep quality are intricate. Consequently, previous research has focused on the mediating or moderating effects of individual variables rather than their combined impact on psychiatric nurses’ sleep quality [28]. This study integrates workplace violence, emotional labour, occupational identity, and sleep quality into a comprehensive model to examine their interrelationships.
The hypotheses for this study are as follows:
Hypothesis 1
There is a significant positive association between workplace violence and poor sleep quality.
Hypothesis 2
There is a significant positive association between workplace violence and surface acting in emotional labour.
Hypothesis 3
There is a significant positive association between workplace violence and deep acting in emotional labour.
Hypothesis 4
There is a significant positive association between surface acting in emotional labour and poor sleep quality surface acting in emotional labour.
Hypothesis 5
There is a significant positive association between deep acting in emotional labour and poor sleep quality.
Anzeige
Hypothesis 6
Surface acting in emotional labour mediate the relationship between workplace violence and poor sleep quality.
Hypothesis 7
Deep acting in emotional labour mediate the relationship between workplace violence and poor sleep quality.
Hypothesis 8
Occupational identity moderates the relationship between workplace violence and surface acting.
Hypothesis 9
Occupational identity moderates the relationship between workplace violence and deep acting.
Anzeige
Methods
Site, setting, and design
This cross-sectional study was conducted in 2022, using a convenience sample of psychiatric nurses working in wards of the Liaoning Mental Health Centre. An online questionnaire was sent to the participating nurses. After providing informed consent, the participants individually completed the questionnaires online according to the instructions. The study included nurses who had been registered for over one year, owned and had access to a mobile phone, and worked in wards.
Population and Sample size calculation
Applying Cohen’s theory of statistical efficacy analysis [29], the minimum sample size required for this study was calculated using the appropriate R language power package. The significance level (α) was set at 0.05, the effect size (f2) was determined as 0.15, and the efficacy power was set at 0.95. The independent variables included age, marital status, educational background, special events, disease status, night shift presence, weekly working hours, workplace violence exposure, surface acting behaviour, deep acting behaviour, and occupational identity. Based on these parameters and considering a potential invalid response rate of 20%, the total number of study participants needed was estimated to be 215. We collected a total of 330 questionnaires. Nurses whose responses did not fit the internal logic of the questionnaire were excluded, and 300 valid questionnaires were actually collected with an effective rate of 90.91% in this study.
To ensure comprehensive and accurate reporting of our study, we used the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for cross-sectional studies [30].
Ethical considerations
The study procedures were reviewed and approved by the Ethics Committee of China Medical University. All participants were provided with information regarding the purpose of the study, and both verbal and written informed consent was obtained after emphasizing that participation was entirely voluntary. The researchers ensured data confidentially and they had the option to withdraw from the study at any time. This study was registered with the Chinese Clinical Trial Registry on 2 August 2022 (registration number ChiCTR2200062347).
Anzeige
Measures
Demographic characteristics
Demographic indicators of the study population included: age; gender; marital status; educational attainment; occurrence of significant life events within the past year such as bereavement, illness in the family, job loss in the family, child’s promotion, or child’s marriage; presence of hypertension or sexual diseases; monthly income level; years of work experience; weekly working hours; night shift work availability.
Sleep quality
Sleep quality was assessed using the Pittsburgh Sleep Quality Index (PSQI), a validated tool developed by Buysse et al. [31]. The PSQI consists of 9 sections and 18 questions. It is widely used to evaluate sleep quality in clinical and research settings. Each item of the scale is rated on a four-point Likert scale ranging from 0 to 3, and the cumulative score for all components yields a total PSQI score (ranging from 0 to 21), with higher scores indicating poorer sleep quality. A meta-analysis conducted on studies involving nurses has reported that a cutoff score of 5 on the PSQI was observed in approximately 62.3% of cases [32]. Therefore, in this study, we adopted a cutoff point of 5 to distinguish between good and poor sleep; that is, nurses were classified as having poor sleep quality if their PSQI scores exceeded 5. The internal consistency reliability coefficient (Cronbach’s alpha) of the PSQI in our study was high at 0.876.
Workplace violence
The 5-item Workplace Violence Scale (WVS), based on the Schat Workplace Violence Scale and adapted to the national context of China, was used to assess the occurrence of workplace violence [33, 34]. The scale encompasses all forms of workplace violence that respondents may have encountered in the past year. It comprises five dimensions: physical assault, emotional abuse, threatening intimidation, verbal sexual harassment, and physical sexual harassment. Each item is scored between 0 and 3, representing frequencies of 0, 1, 2–3 times, or ≥ 4 times, respectively; higher scores indicate more frequent occurrences of workplace violence. The WVS is widely employed in medical, nursing, and psychiatric care settings with demonstrated reliability and exhibited a Cronbach’s alpha coefficient of 0.891 in this study.
Emotional labour
A 14-item three-dimensional Emotional Labor Scale was developed by Diefendorff et al. [35]. The scale measures surface acting, deep acting, and natural flow as dimensions of emotional labour strategy. For this study, only the surface- and deep-acting dimensions were assessed using Questions 1–7 and 8–11, respectively. Responses were scored on a five-point Likert scale ranging from 1 (never) to 5 (always), with higher scores indicating greater levels of a specific type of emotional labour. This scale has been widely used in nurses with good reliability and validity [36]. Cronbach’s alpha coefficients for the surface- and deep-acting subscales in this study were 0.933 and 0.821, respectively.
Occupational identity
The Occupational Identity Scale, developed by Tyler and McCallum [37, 38], was translated into Chinese by Professor Tsai Chang-juan of Chaoyang University of Science and Technology in Taiwan for studying occupational identity among Chinese workers. The scale comprises 10 items rated on a five-point Likert scale, with scores ranging from 1 (not at all) to 5 (completely); higher scores indicate greater occupational identity. In this study, Cronbach’s alpha for the scale was 0.951.
Data analysis
SPSS Statistics 21.0 (IBM, Asia Analytics Shanghai) was used for the descriptive and statistical analyses. Descriptive statistics were conducted on the general demographics of the sample, poor sleep quality scores, emotional labour (surface and deep) scores, frequency of workplace violence, and occupational identity using means, standard deviations, or relative numbers. Independent sample t-tests and one-way ANOVA were used to compare differences in sleep quality and emotional labour (surface and deep) between psychiatric nurses with different demographics. Pearson’s correlation analysis was used to continuously test the data. Age was included as a continuous variable and we performed a test of covariance between variables. The mediating effect of emotional labour between workplace violence and sleep quality was examined using hierarchical multiple linear regression. The moderating effect of professional identity on the relationship between workplace violence and emotional labour was also investigated.
The SPSS plug-in PROCESS macro (version 3.0 by Andrew F. Hayes) was used for data analysis. PROCESS Model 4 was used to investigate the mediating role of surface acting and deep acting in emotional labour in the relationship between workplace violence and poor sleep quality. Additionally, we examined the moderating role of occupational identity between workplace violence and surface acting using PROCESS Model 1. Subsequently, a moderated mediator model (PROCESS Model 7) was tested with workplace violence as the independent variable, occupational identity as the moderator variable, surface and deep acting as the mediators, and poor sleep quality as the dependent variable [39]. Subsequently, simple slope analyses were conducted to further investigate the impact of workplace violence on poor sleep quality across different levels of occupational identity. Before model testing, the scores for workplace violence, emotional labour (surface and deep acting), sleep quality, and occupational identity were standardised. Additionally, continuous variables were centred before examining the moderating effect. P-values < 0.05 were considered statistically significant.
Results
Descriptive statistics and univariate analysis
The study results revealed a sleep quality score of 6.87 ± 4.58 and a 55% prevalence of poor sleep quality among psychiatric nurses. They had a mean score of 11.81 ± 3.49 for surface acting and 19.12 ± 6.73 for deep acting. Table 1 shows that the majority of nurses are female (66.33%) and about two thirds of the participants were married (72.33%). In addition, more than half of the nurses had more than 10 years of work experience. Nurses who were unmarried, divorced, separated, or widowed; hold a bachelor’s degree or higher; have experienced a significant event, such as the death of a relative or marriage of their child, within the past month; suffer from an underlying chronic illness such as hypertension; and work more than 40 h per week or during night shifts exhibited higher scores for surface acting and PSQI indicating poor sleep quality. Furthermore, nurses who hold a bachelor’s degree, have experienced a significant event, or suffer from an underlying chronic illness exhibited higher scores for deep acting (Table 1).
Table 1
Descriptive statistics and univariate analysis
Variables | n (%) | Sleep quality | Surface acting | Deep acting | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
M ± SD | t/F-value | CI | p-value | M ± SD | t/F-value | CI | p-value | M ± SD | t/F-value | CI | p-value | ||
Gender | 0.993 | (-0.545,1.655) | 0.321 | 1.682 | (-0.244,2.984) | 0.096 | 0.384 | (-0.683,0.995) | 0.714 | ||||
Male | 101(33.67%) | 6.50 ± 4.55 | 18.21 ± 6.62 | 11.70 ± 3.17 | |||||||||
Female | 199(66.33) | 7.06 ± 4.59 | 19.58 ± 6.76 | 11.86 ± 3.64 | |||||||||
Marital status | 2.082 | (0.052,2.364) | 0.041 | 2.057 | (0.120,3.521) | 0.036 | 0.961 | (-0.418,1.352) | 0.300 | ||||
Married/cohabiting | 217(72.33%) | 6.54 ± 4.59 | 18.61 ± 6.59 | 11.68 ± 3.30 | |||||||||
Unmarried/divorced/separated/widowed | 83(27.67%) | 7.75 ± 4.46 | 20.43 ± 6.96 | 12.14 ± 3.93 | |||||||||
Educational level | -2.009 | (-2.247, -0.124) | 0.029 | -5.496 | (-6.164, -2.884) | < 0.001 | -2.563 | (-2.007, -0.246) | 0.012 | ||||
Graduate and below | 82(27.33%) | 6.01 ± 3.89 | 15.83 ± 6.31 | 10.99 ± 3.36 | |||||||||
Bachelor’s degree and above | 72.67(72.67%) | 7.20 ± 4.78 | 20.35 ± 6.48 | 12.11 ± 3.49 | |||||||||
Significant event | -6.631 | (-4.727, -2.552) | < 0.001 | -3.214 | (-4.340, -0.968) | 0.002 | -2.815 | (-2.092, -0.341) | 0.007 | ||||
Yes | 83(27.67%) | 9.51 ± 4.24 | 21.04 ± 6.24 | 12.69 ± 3.29 | |||||||||
No | 217(72.33%) | 5.87 ± 4.30 | 18.38 ± 6.78 | 11.47 ± 3.51 | |||||||||
Chronic illness | -6.626 | (-4.391, -2.379) | < 0.001 | -3.466 | (-4.217, -1.170) | 0.001 | -2.679 | (-1.867, -0.277) | 0.008 | ||||
Yes | 125(41.67%) | 8.85 ± 4.59 | 20.69 ± 6.70 | 12.43 ± 3.33 | |||||||||
No | 175(58.33%) | 5.46 ± 4.01 | 17.99 ± 6.55 | 11.36 ± 3.53 | |||||||||
Monthly income | 1.428 | (-0.320, 1.852) | 0.166 | 1.475 | (-0.358,2.746) | 0.131 | 0.484 | (-0.602,1.010) | 0.619 | ||||
RMB 5,000 and below | 123(41.00%) | 7.33 ± 4.97 | 19.82 ± 7.28 | 11.93 ± 3.80 | |||||||||
RMB 5,000 and above | 177(59.00%) | 6.56 ± 4.26 | 18.63 ± 6.30 | 11.72 ± 3.26 | |||||||||
Years of work experience | 1.285 | 0.278 | 2.325 | 0.1 | 0.049 | 0.952 | |||||||
1–5 years | 43(14.33%) | 6.91 ± 4.63 | 20.28 ± 6.84 | 11.79 ± 4.08 | |||||||||
6–10 years, | 85(28.33%) | 7.52 ± 4.84 | 19.98 ± 6.57 | 11.91 ± 3.61 | |||||||||
> 10 years | 172(57.33%) | 6.55 ± 4.42 | 18.40 ± 6.73 | 11.76 ± 3.28 | |||||||||
Weekly working hours | -5.754 | (-3.959, -1.888) | < 0.001 | -2.399 | (-3.423, -0.351) | 0.016 | -0.322 | (-0.933,0.672) | 0.749 | ||||
≤ 40 h | 173(57.67%) | 5.64 ± 3.91 | 18.32 ± 6.54 | 11.75 ± 3.55 | |||||||||
> 40 h | 127(42.33%) | 8.56 ± 4.89 | 20.2 ± 6.87 | 11.88 ± 3.40 | |||||||||
Night shift | -4.774 | (-3.506, -1.611) | < 0.001 | -2.996 | (-4.031, -0.859) | 0.003 | -1.444 | (-1.432,0.229) | 0.155 | ||||
Yes | 196(65.33%) | 7.76 ± 4.87 | 19.96 ± 6.55 | 12.02 ± 3.53 | |||||||||
No | 104(34.367%) | 5.20 ± 3.39 | 17.52 ± 6.82 | 11.41 ± 3.39 | |||||||||
Correlation analysis
Psychiatric nurses’ age was negatively associated with violence and poor sleep, and positively associated with occupational identity. In other words, older psychiatric nurses reported less workplace violence, improved sleep quality, and greater occupational identity than younger nurses. Workplace violence was positively associated with poor sleep quality, surface acting, and deep acting. Occupational identity was negatively correlated with workplace violence, poor sleep quality, and surface acting (Table 2).
Table 2
Correlation analysis of influencing factors of sleep quality
Variables | M ± SD | Age(CI) | Workplace violence(CI) | Sleep quality(CI) | Surface acting(CI) |
|---|---|---|---|---|---|
Age | 38.78 ± 9.82 | ||||
Workplace violence | 2.21 ± 3.53 | -0.226*** (-0.308, -0.131) | |||
Poor sleep quality | 6.87 ± 4.58 | -0.173***(-0.263, -0.072) | 0.450***(0.332,0.559) | ||
Surface acting | 11.81 ± 3.49 | -0.280***(-0.386, -0.168) | 0.322***(0.212,0.424) | 0.272***(0.162,0.374) | |
Deep acting | 19.12 ± 6.73 | -0.131*(-0.238, -0.019) | 0.256***(0.138,0.361) | 0.122*(0.010,0.226) | 0.719***(0.641,0.794) |
Analysis of mediating effect
The variance inflation factor for this study was ≤ 2.484 with a minimum tolerance of 0.403, which is greater than 0.1, indicating no effect of covariance problems on the results [40, 41].
First, the model included age, marital status, education, significant events, medical conditions, night shifts, and hours worked per week as control variables. Second, workplace violence was introduced as an independent variable. After controlling for the effects of the control variables, including age and marital status, workplace violence significantly impacted sleep quality, explaining 9.5% of the variance. Third, surface and deep acting were added to the model as mediators, accounting for an additional 1.2% of the variance in sleep quality. The standardised regression coefficient for workplace violence decreased slightly from 0.328 in step two to 0.325 in step three but remained statistically significant. Therefore, it is suggested that surface acting partially mediates the relationship between workplace violence and poor sleep quality, whereas deep acting has a suppressive effect (Table 3).
Table 3
Hierarchical regression analysis of surface acting, deep acting and workplace violence on poor sleep quality
Variables | Step 1 (β) | Step 2 (β) | Step 3 (β) |
|---|---|---|---|
Age (years) | -0.076 | -0.010 | 0.003 |
Marital status | -0.118* | -0.108* | -0.104 |
Educational level | 0.045 | 0.044 | 0.025 |
Significant event | 0.221*** | 0.200*** | 0.202*** |
Chronic illness | 0.266*** | 0.218*** | 0.212*** |
Night shift | 0.016 | 0.023 | 0.027 |
Weekly working hours | 0.215*** | 0.164** | 0.147** |
Workplace violence | 0.328*** | 0.325*** | |
Surface acting | 0.153* | ||
Deep acting | -0.149* | ||
F | 16.278*** | 21.920*** | 18.300*** |
Adjusted R2 | 0.263 | 0.359 | 0.367 |
△R2 | 0.281 | 0.095 | 0.012 |
The mediation analysis is shown in Fig. 1; the indirect impact of workplace violence on sleep quality was found to be significant through surface acting (a1*b1 = 0.0349, 95% confidence interval [CI] = [0.0017, 0.0790]). Surface acting plays a partial mediating role between workplace violence and poor sleep quality, with an effect size of |a1 * b1 / c| = 10.67%. Additionally, the indirect effect of workplace violence on sleep quality was significant through deep acting (a2*b2=-0.0327, 95% CI = [-0.0713, -0.0029]), and deep acting exhibits a masking effect with an effect size of |a2*b2/c| = 9.99%. Therefore, without controlling for deep acting, the difference in sleep quality between nurses who experienced more workplace violence and the ones who experienced less was masked; however, when controlling for deep acting, the difference widened.
Fig. 1
Mediating Effect analysis
×
![]()
Moderating effect of occupational identity between workplace violence and surface and deep acting
First, the control variables age, marital status, education, significant events, medical conditions, night shifts, and hours worked per week were included in the model. Second, the independent variable of workplace violence and moderating variable of occupational identity were entered. The interaction term significantly affected surface acting (β = 0.112, p < 0.05), suggesting that occupational identity may moderate the relationship between workplace violence and surface acting (Table 4). However, there was no moderating effect between workplace violence and deep acting (Table 5).
Table 4
Hierarchical regression analysis of workplace violence and occupational identity on surface acting
Variables | Step 1 (β) | Step 2 (β) | Step 3 (β) |
|---|---|---|---|
Age (years) | -0.166* | -0.122 | -0.122 |
Marital status | -0.088 | -0.064 | -0.075 |
Educational level | 0.221*** | 0.218*** | 0.219*** |
Significant event | 0.087 | 0.067 | 0.073 |
Chronic illness | 0.159** | 0.103 | 0.093 |
Night shift | -0.018 | -0.037 | -0.033 |
Weekly working hours | 0.077 | 0.035 | 0.021 |
Workplace violence | 0.189** | 0.156** | |
Occupational identity | -0.182** | -0.209*** | |
Workplace violence × Occupational identity | 0.112* | ||
F | 9.015*** | 10.907*** | 10.349*** |
Adjusted R2 | 0.158 | 0.230 | 0.238 |
△R2 | 0.178 | 0.075 | 0.011 |
Table 5
Hierarchical regression analysis of workplace violence and occupational identity on deep acting
Variables | Step 1 (β) | Step 2 (β) | Step 3 (β) |
|---|---|---|---|
Age (years) | -0.077 | -0.033 | -0.040 |
Marital status | -0.065 | -0.057 | -0.048 |
Educational level | 0.096 | 0.095 | 0.095 |
Significant event | 0.107 | 0.092 | 0.087 |
Chronic illness | 0.118 | 0.084 | 0.092 |
Night shift | 0.008 | 0.010 | 0.007 |
Weekly working hours | − 0.037 | -0.071 | -0.060 |
Workplace violence | 0.215*** | 0.243*** | |
Occupational identity | -0.020 | 0.003 | |
Workplace violence × Occupational identity | -0.093 | ||
F | 2.777** | 3.796*** | 3.672*** |
Adjusted R2 | 0.040 | 0.078 | 0.082 |
△R2 | 0.062 | 0.043 | 0.007 |
The relationship between workplace violence and surface role was significantly different at low (effect = 0.116, p = 0.067, 95% CI = [-0.0081,0.2406]), middle (effect = 0.207, p < 0.001, 95% CI = [-0.0996,0.3149]), and high (effect = 0.298, p < 0.001, 95% CI = [0.1470,0.4494]) occupational identity levels. As Fig. 2 shows, the positive correlation between workplace violence and surface acting becomes significant as occupational identity level increases.
Fig. 2
Simple slope analysis
×
![]()
Moderated mediation accounted for 38.77% of the variance in poor sleep quality among psychiatric nurses. Notably, the interaction between workplace violence and occupational identity significantly influenced surface acting (β = 0.091, p < 0.05). However, the two moderating mediating effects examined in this study were not confirmed (Table 6).
Table 6
Moderated mediation effects
Relationship | β | P | 95% CI | |
|---|---|---|---|---|
Workplace violence →Surface acting | 0.207 | < 0.001 | (0.0996 0.3149) | |
Workplace violence →Deep acting | 0.194 | < 0.010 | (0.0747, 0.3138) | |
Workplace violence→Poor sleep quality | 0.325 | < 0.001 | (0.2260,0.4241) | |
Surface acting→Poor sleep quality | 0.156 | 0.036 | (0.0104,0.3019) | |
Deep acting→Poor sleep quality | -0.151 | 0.028 | (-0.2848,0.0164) | |
Occupational identity →Surface acting | -0.207 | < 0.001 | (-0.3149, -0.0983) | |
Occupational identity→Deep acting | 0.003 | 0.9669 | (-0.1177,0.1228) | |
Workplace violence×Occupational identity →Surface acting | 0.091 | 0.0406 | (0.0039,0.1788) | |
Workplace violence×Occupational identity →Deep acting | -0.076 | 0.1233 | (-0.1733,0.0208) | |
Index of moderated mediation = 0.0143 | (-0.0131,0.0395) | |||
Index of moderated mediation = 0.0115 | (-0.0113,0.0421) | |||
R2 | 0.3877 | |||
Discussion
This sleep quality among psychiatric nurses is lower than the national adult norm in China, indicating poorer sleep quality compared with the general population, which aligns with Lyu et al.’s [8] findings. Univariate analysis revealed that nurses who were unmarried, divorced, separated, or widowed, and those with a bachelor’s degree or higher were at an elevated risk of experiencing poor sleep quality. Additionally, factors such as encountering significant events in the past year, suffering from chronic conditions such as hypertension and diabetes, working more than 40 h per week, and engaging in night shifts further contributed to this heightened risk. Nurses must enhance their awareness of sleep health and improve their sleep habits by adhering to regular bedtimes and refraining from using mobile phones before going to bed. Managers can play a crucial role in assisting nurses by fostering harmonious relationships between colleagues and providing emotional support, specifically for individuals who are unmarried, divorced, or widowed. Simultaneously, hospitals should strive to optimise nurses’ working hours and night-shift schedules.
The occurrence of workplace violence can result various forms of physical harm among psychiatric nurses, including pain, limb injuries, and eating disorders [42]. The incidence of sleep disturbances, including difficulties initiating and maintaining sleep, frequent awakenings, and the occurrence of nightmares, has been observed to increase in individuals exposed to workplace violence. Furthermore, there is an elevated risk of nurses developing psychological issues, including depression, anxiety, insomnia, post-traumatic stress disorder (PTSD) symptoms, intense fear, and a diminished sense of security [43]. Our findings suggest that workplace violence significantly impacts sleep quality in psychiatric nurses, which is consistent with previous research [44].
In comparison to the scores of surface acting and deep acting in other healthcare workers in previous studies, the mean scores of psychiatric nurses in surface acting were found to be lower, while the mean scores of deep acting were higher [16, 45]. This suggests that psychiatric nurses may engage in less surface acting and more deep acting than other healthcare professionals. This may also be associated with higher levels of empathy among psychiatric nurses [46]. However, in our study, surface acting partially mediates the relationship between workplace violence and poor sleep quality. The phenomenon of insomnia occurs when individuals suppress negative emotions for the purpose of surface acting [47]. Prior research has demonstrated that surface acting is significantly and positively correlated with stress and burnout, and higher levels of stress and burnout have been identified as predictors of lower sleep quality [14, 48]. Incidents of violence in the workplace deplete nurses’ physical and emotional resources, resulting in physical harm and triggering negative emotions such as anger and depression [49]. Nurses are expected to suppress these negative emotions to meet organisational expectations and engage in surface acting, further depleting their resources [16]. Failure to promptly replenish these resources can lead to caregiver fatigue and stress responses [50], ultimately resulting in a substantial decline in individual sleep quality [51].
Psychiatric nurses are recognised as healthcare professionals who demonstrate a high level of empathy towards patients with mental disorders [46]. Following workplace violence, psychiatric nurses may be more willing to understand the reasons behind a patient’s violent behaviour and think from the patient’s perspective [46]. This emotional connection and understanding may facilitate nurses to effectively manage their own emotions, think positively and adjust, understand and empathise, and act accordingly at a deeper level.
Findings suggest that nurses’ poor sleep quality may be affected by deep acting in emotional labour and that deep acting plays a suppression effect in the impact of workplace violence on poor sleep quality. Engaging in deep acting positively impacts sleep quality and helps mitigate the negative effects of workplace violence on sleep quality, which aligns with Ma et al.’s findings [52]. According to the resource conservation theory, deep acting can authentically express an individual’s self-feelings, reduce perceived stress, and harvest positive psychological resources such as a sense of personal achievement [53, 54]. Deep acting is beneficial to individuals because it helps them acquire or replenish external resources. It positively affects individuals when they receive more resources than they expend [14]. Therefore, hospital managers should encourage nurses to frequently engage in deep acting during emotional work to ensure optimal sleep. This can be achieved by fostering positive mental states and concentration skills to support nurses in their practice of deep acting [45].
This study highlights a significant finding regarding the impact of occupational identity on the relationship between workplace violence and surface acting. Specifically, nurses with a stronger sense of occupational identity tend to engage in more surface acting when confronted with workplace violence than those with a weaker sense of occupational identity. Occupational identity is a valuable asset, linked to higher job satisfaction [55] and increased work engagement [56]. Previous studies have primarily focused on the positive impact of a strong occupational identity on individual health and job performance [56]. Individuals with a strong occupational identity may experience an imbalance between their giving and receiving experiences [57]. This imbalance, particularly among nurses with high occupational identity, leads to increased negative emotions and decreased work engagement [58]. Nurses’ reluctance to invest time and effort in transitioning from surface to deep acting stems from the convenience and comfort of surface acting. Additionally, psychiatric settings are characterised by a high prevalence of workplace violence, significantly diminishing nurses’ emotional engagement with their work [17]. Therefore, organisations should intervene comprehensively in all processes related to workplace violence to effectively minimise and address its underlying causes.
Limitations
This study has some limitations. First, the design of this study was cross-sectional, which restricted our ability to establish causality. Second, the generalisability of our findings may be limited because the study population was derived solely from a single mental health centre. To address this concern, we plan to conduct a multicentre study with a larger sample size. Finally, it is important to acknowledge that information bias cannot be completely eliminated because of the reliance on self-administered questionnaires for data collection.
Conclusion
Nurses employed in psychiatric settings often experience suboptimal sleep quality, which is significantly influenced by workplace violence, surface acting, and deep acting. Moreover, the association between workplace violence and poor sleep quality is partially mediated by surface acting and obscured by deep acting. Occupational identity serves as a positive moderator linking workplace violence to surface acting, suggesting that an increase in occupational identity intensifies nurses’ engagement in surface acting when exposed to workplace violence. Hospital managers should acknowledge the emotional, physical, and psychological burdens of workplace violence on the mental health of nurses and implement appropriate measures to establish a secure work environment while minimising and preventing incidents of workplace violence. Following an incident of workplace violence, organisations should provide comprehensive consultation and actively encourage nurses to engage in thorough emotional processing to augment their existing resources. Additionally, organisations must offer nurses appropriate emotional support and compassionate care that can effectively enhance their sleep quality and promote physical and psychological well-being.
Acknowledgements
We would like to extend sincere gratitude to all those who contributed to this research. Their collaboration and unwavering dedication were instrumental in enabling the realisation of this research endeavour. Additionally, we acknowledge with gratitude all individuals who provided professional advice throughout each stage of this extensive investigation.
Declarations
Ethics approval and consent to participate
The study procedures were reviewed and approved by the Ethics Committee of China Medical University. All participants provided written informed consent. The researchers ensured data confidentially.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
