Design and sampling

This was a cross-sectional survey study. Convenience sampling was employed to recruit university NSs studying in 3-year higher diploma and 4-year bachelor programmes, in classroom settings, from March to June 2012. The NSs were informed, verbally and through a written information sheet, about their voluntary participation; their consent was implied if they returned the completed questionnaire. In addition, they were assured that their decision to participate or not would not affect their academic results. No incentive was given for participation.

Instrument

A questionnaire named ‘Clinical Violence towards NSs’ was adapted based on the literature review.16 28 29 Permission to use the questionnaire was obtained from the authors. Substantial modifications were made to meet the study objectives. The definition of clinical violence used by the International Labour Organization and coorganisations16 was adopted and was stated at the beginning of the questionnaire.

The questionnaire had three sections. Section 1 consisted of 11 items to collect personal information (age, gender, programme and year of study), and respondents’ perceptions about clinical violence. They were asked to rate their susceptibility to violence in their clinical placements, the extent of their concern about clinical violence, whether they perceived it to be a part of the nursing job and their satisfaction with the training provided by their study programmes. In addition, they were asked to identify the workplace factors contributing to clinical violence, such as patients or visitors under the influence of alcohol or drugs. Furthermore, they were asked to indicate if they had witnessed or experienced physical violence and/or verbal abuse.

Sections 2 and 3 of the questionnaire covered the experiences of physical violence and verbal abuse, respectively, in the clinical placement. The NSs were required to complete either or both sections if they had experienced physical violence and/or verbal abuse. Each section contained 32 items covering four areas: (1) information about the violence (either physical or verbal) experienced, including frequency during the study period and the prior 12 months, place of the occurrence of violence, shift involved and the perpetrators; (2) actions taken in responding to the violence; (3) reporting behaviours; and (4) impact of the violence on personal emotions, clinical performance, how much they were disturbed by the violence and their intention to leave the nursing profession. The impact on personal emotions was assessed by 10 items asking their feelings about the incident, such as frustration, anger, fear, irritability, sadness, headache, difficulty in sleeping, shame, depression or low self-esteem. The effect on clinical performance was evaluated by four items asking if they had lost confidence, had difficulty concentrating, provided poor nursing care to patients or experienced decreased grades for clinical placement. Furthermore, the participants were asked how disturbed they were by the violence, using four items; repeated, disturbing thoughts or images of the violence; avoiding thinking about the violence; being ‘super-alert’; and feeling tired and needing to make an effort to do everything. These three subscale impact items were rated using 5-point Likert scales (ie, 1=not at all to 5=extremely). The average of each subscale was used for the data analyses. Higher scores indicated greater impact. Last, the participants were asked to evaluate their intentions to leave the nursing profession before and after the violence, using a 5-point Likert scale (ie, 1=never thought to 5=always thought).

The study questionnaire was validated by a panel of four local and overseas experts in the field of clinical violence and occupational health. The content validity index was 0.98, which was considered acceptable.30 Furthermore, the reliability of the questionnaire was tested using the 2-week test–retest method with 30 NSs. The reliability coefficient was 0.73, which was also considered acceptable.30

Data analysis

The data were analysed using IBM SPSS Statistics V.23. Descriptive statistics were used to present the frequencies, percentages, means and SD of the variables under study. Missing data were not replaced because the maximum percentage of missing data for the study variables was 0.07%. χ² and independent t-tests were used to examine the factors (personal and workplace), associated with physical violence and/or verbal abuse. For the participants who had experienced both physical violence and verbal abuse, dependent t-tests and Wilcoxon signed-ranks test were used to determine the differences in characteristics (such as responses to the incidents and reasons for not reporting them formally) between physical violence and verbal abuse, as well as changes in their intentions to leave before and after the experiences of clinical violence. A p value of <0.05 was considered statistically significant.

Patient and public involvement

Patients were not involved in this study.

Characteristics of the participants and associated factors of clinical violence

Table 1 summarises the participants’ characteristics. The gender ratio of 70.4% (female) to 29.6% (male) was consistent with the ratio of the overall numbers of students enrolled in the programmes (68% to 32%). The distribution of NSs across different years of study was an average of 25% per year. In general, about 50% of the NSs perceived low susceptibility to and low concerns about clinical violence. These perceptions might have been due to their thinking that clinical violence is not a part of the nursing job (73.3%, n=737). On the other hand, close to two-thirds of the NSs considered their training for coping with violence was not adequate. Patient-related factors and heavy workloads of nursing staff were frequently stated as the factors associated with clinical violence.

However, the participants’ characteristics changed if they had experienced either physical violence, verbal abuse or both. For instance, their perceived susceptibility to clinical violence (p<0.001; χ² 60.59) and concerns about it (p=0.002; χ² 20.44) increased incrementally; the lowest ratings were made by those who had not experienced any physical violence or verbal abuse, followed by those who had been exposed to physical violence only, those who had suffered verbal abuse only and then those having both experiences. The perceived associated factors of clinical violence also changed based on the NSs’ experiences. Those experiencing both physical violence and verbal abuse were more likely to perceive confused patients as the associated factor. On the other hand, those experiencing only physical violence were more likely to consider high patient volume as the associated factor. In addition, the NSs receiving only verbal abuse were more likely to perceive staff shortages as the factor. Interestingly, those without any experience of either type of violence were more likely to identify alcohol, drug influence and uncaring nursing behaviours as the factors. Furthermore, as expected, year 4 NSs had experienced more clinical violence than those in other years (p<0.001; χ² 233.17). Tukey’s post hoc tests further indicated that those without any clinical violence experience were younger than those with such experience (F₃₉₉₁ 31.78; p<0.001).

Prevalence and perpetrators of physical violence and verbal abuse

Table 2 shows the comparisons of the prevalence and the perpetrators of clinical violence. Of the 1017 NSs, 168 (16.5%) and 311 (30.6%) had personally experienced physical violence and verbal abuse, respectively, in their clinical placements. Of these, 100 (9.8%) had experienced both physical violence and verbal abuse. Thus, a total of 379 participants (37.3%) had experienced clinical violence during their nursing studies. However, the NSs had witnessed more physical violence (25.5%, n=259), verbal abuse (43.9%, n=446) or both (17.6%, n=179) than they had actually experienced personally. Furthermore, it was alarming to observe that 4.0% (n=15) of the participants had experienced verbal abuse all the time (ie, almost every day during clinical placement). Six (1.6%) had experienced physical violence with physical injuries requiring formal treatment.

For the 100 participants who had experienced both physical violence and verbal abuse, the frequency of occurrence of verbal abuse was more than that of physical violence (p<0.001). Patients’ relatives, university supervisors, hospital clinical instructors and ward supervisors were more significantly identified as the perpetrators of verbal abuse than of physical violence.

Responses to and impacts of physical violence and verbal abuse

Table 3 shows that most participants did not take action about the clinical violence or formally report the incidents. Their reasons for not reporting were mainly because they thought it was not important or useless to do so, they did not know who to report to or no one encouraged them to report the incident. The 100 participants who had experienced both physical violence and verbal abuse responded to physical violence and verbal abuse in significantly different ways. Those who had experienced verbal abuse were more likely not to take action, to ask the perpetrators to stop, to try to defend themselves physically or to report the incident, but they were more likely to tell their friends or family (p<0.05). They perceived that physical violence could be more preventable than verbal abuse (p<0.001). Although none of the verbal abuse led to formal treatment, there was significantly more sick leave taken after verbal abuse than after physical violence experiences (p<0.05). The sick leave lasted from 1 to 10 days. Furthermore, the negative effects on their personal feelings and clinical performances and the extent to which they were disturbed by the clinical violence were significantly greater for verbal abuse than for physical violence (p<0.05). The intention to leave the nursing profession after the clinical violence was consistently higher than before its occurrence for participants who had experienced either physical violence, verbal abuse or both (p<0.001).

Prevalence of clinical violence

Our study showed that, while 37.3% of the NSs had experienced clinical violence during their nursing studies, the prevalence of verbal abuse (30.6%) was significantly greater than that of physical violence (16.5%). This indicates that the overall clinical violence and verbal abuse rate was moderately high for these NSs. Our findings were, to some extent, comparable to those of studies conducted in other countries, where overall prevalence rates have been reported as 34% in Italy,14 35.3% in Iran,15 42.2% in UK2 and 50.3% in Turkey.13 However, several studies reported much higher prevalence of verbal abuse of NSs, such as 91.6% in Turkey,1376% in Italy,14 73.3% in Iran15 and 45.1% in the UK.11 The relatively higher rate in the other studies could be explained partly by the differences in the definitions of clinical violence and socioeconomic cultural variations in the studied populations. For instance, Tee et al,2 in the UK, included racism as a form of abuse. Furthermore, because of cultural values and norms, Chinese NSs are likely to be more obedient and respectful to their seniors,24 which may potentially result in lower rates of clinical violence specifically by hospital and/or university staff.

Perpetrators and associated factors of clinical violence

We found that the perpetrators of the verbal abuse experienced by the NSs were predominantly patients (66.8%), followed by hospital staff (29.7%), university supervisors (13.4%) and patients’ relatives (13.2%). Patients (91.0%) were the greatest perpetrators of physical assaults. Our results were largely in agreement with the studies of Ferns and Meerabeau11 and Magnavita and Heponiemi14 but were inconsistent with others.2 15 For instance, Tee et al 2 found that British NSs were confronted more frequently by hospital staff, including nurses, hospital care assistants and managers (31.1%) and less frequently by patients and relatives (4.9%–1.2%, respectively). Despite nursing being caring profession, it is a great concern that there is vertical violence inflicted by hospital staff and university supervisors/teachers. However, the reasons for such vertical violence are not well understood. Future research is necessary to elucidate the contributing factors for such vertical clinical violence. Besides, our findings confirmed that it is not uncommon for NSs to encounter verbal and physical abuse from patients and their relatives. Future research is also needed to understand the characteristics of perpetrators, victims and organisations related to clinical violence towards NSs, and to study the relationships between these variables to elucidate appropriate tailored initiatives and intervention approaches to mitigate workplace violence. Research-based knowledge about the causes and escalating nature of violence incidences would facilitate the planning of interventions.31

An interesting result was found in our study: there were significant differences between NSs with and without clinical violence experiences on perceived susceptibility of, concern about and associated factors with clinical violence. NSs with clinical violence experiences commonly believed that the reasons for clinical violence were the hospital system (such as staff shortages and high patient volume) and confused patients, while those without such experiences blamed NSs’ uncaring attitudes or patients/visitors under the influence of drugs/alcohol. These differences concur with the health belief model.32 A person’s perceived susceptibility and assessment of the severity of an event (such as clinical violence) are affected by his/her knowledge and experience of that event. Our study found that NSs who had experienced clinical violence perceived themselves to be more susceptible and the violence to be more severe than did their counterparts. As well, the majority of NSs (74.6%) recognised that their training about clinical violence was inadequate and inappropriate. The results of the study echoed the claims in the existing literature that clinical violence towards nurses is associated with understaffing, patient-risk factors and lack of assault management training.33 Thus, incorporating training programmes for NSs in violence prevention and management can be a fundamental strategy to decrease clinical violence. Although all graduating NSs in our university do undergo violence prevention and management training, our study results can inform the revision of the training programme to include how to assess and communicate with confused patients in an understaffed clinical environment. Future study is also needed to examine whether such training would enhance NSs’ competence in managing clinical violence. According to the Framework Guidelines for addressing workplace violence in the health sector,16 the areas that should be considered include early recognition of escalating behaviours and situations, de-escalating techniques in interpersonal and communication skills and enhanced knowledge about medications to control patients’ agitation better. Initiating antiviolence policies, together with crisis interventions, to reduce workplace violence, may play a critical role in violence prevention strategies as well as improvements to workplace safety. Appropriate policies against workplace violence, with priorities given to work ethics, safety, mutual respect, tolerance, equal opportunity and cooperation, should be developed and implemented to address workplace violence.16 Research evidence has demonstrated that early intervention with verbal-escalation conducted by multidisciplinary assault reduction teams can be effective in resolving violent incidences and decreasing nursing staff injuries by 47% in hospital settings.34 The multidisciplinary assault reduction team is formulated by the nursing supervisor, security personnel, the primary physician and nurse and others involved in the patient’s direct care. All team members have undergone specialised verbal de-escalation training.

Impact of clinical violence

Notably, clinical violence has detrimental effects on NSs. It is noteworthy that all of these negative effects on emotion, clinical performance and the extent to which the respondents were disturbed by the incidents, were significantly greater for verbal abuse than for physical violence. Additionally, clinical violence deterred our future nurses from staying in the profession. Their intention to leave the nursing profession after experiencing clinical violence was significantly higher than it was before the experience (p<0.001). Moreover, verbal abuse resulted in students’ absenteeism from work (9.9%). In our study, the NSs who had experienced verbal abuse (78.9%) were more likely than those who had experienced physical violence (73.5%) to take no action or to pretend the violence had not happened (table 3). These incidences were rarely reported because the students felt that they were either trivial or that reporting would be futile. Our results were congruent with other studies.2 13 14 Violence against NSs not only causes psychological harm but also can affect their clinical performances, compromising the quality of patient care; more importantly, it may lead them to abandon their profession as the result of the violence. Our findings highlight the gaps in current strategies and interventions available to alleviate clinical violence, particularly to address verbal abuse from authority figures so as to protect NSs from being victims.

According to the International Labour Organization’s framework guidelines for addressing workplace violence in the health sector,16 clinical violence reduction initiatives and strategies are essential and can be presented by both individual and system approaches in hospital settings. It is suggested that assertiveness empowerment training and self-defence should be provided for NSs as individual-focused interventions. To improve coping with workplace violence, general well-being should be promoted by maintaining physical fitness and emotional stability. As a caring profession, it is necessary at management level (both educational and clinical) to establish protocols for reporting, documenting and responding to incidents of violence. Increasing NSs’ awareness about how and where to report without fear of criticism or reprisal would help to unveil violence incidents and tailor appropriate preventive and management strategies. Reported cases can be referred to counselling services for emotional support and improved coping strategies.