Workplace violence experienced by nurses: associated factors and impact on mental health

¹Hedi Chaker University Hospital, University of Sfax, Sfax, Tunisia

^&Corresponding author
Nada Kotti, Hedi Chaker University Hospital, University of Sfax, Sfax, Tunisia

Introduction    

Health care workers are among the groups most experiencing aggressive behavior at work. The National Institute for Occupational Safety and Health notes that hospital employees are at high risk for experiencing violence in the workplace, especially nurses [1]. This category takes a leading role within health-care teams, and often works at the front line interacting with all kinds of patients and has direct contact with their families and relatives. WPV is known as one of the most challenging issues in negative nursing work environments. Groenewold et al. (2017) analyzed surveillance data about injury due to workplace violence from 2012 to 2015 in the US. They found that nurses have the highest workplace violence injury rates of all health professionals, finding that the nurses in their study faced 1.7 times the risk of experiencing injury due to workplace violence than personnel not involved in patient care [2]. A meta-analysis of 136 international research studies, conducted in Asian, European, Anglo and Middle East regions demonstrated that verbal abuse is the most frequent type of violence reported by nurses [3]. This situation may be highly distressing for nurses and entails a series of negative consequences, since verbal and physical violence against workers is associated with decreased job satisfaction, increased occupational strain, and poor patient care outcomes. The nurse may also respond to aggression with absenteeism from work, changing jobs, or leaving the nursing profession altogether [4]. This study aimed to determine the prevalence rate of WPV against nurses in Tunisian university hospitals, to identify associated factors and to evaluate the impact on nurses´ mental health.

Methods     

Setting and participants

This was a cross-sectional survey conducted from April to July 2019 at two public hospitals belonging to the Ministry of health in Sfax, Tunisia: Hedi Chaker University Hospital and Habib Bourguiba University Hospital. A self-administered questionnaire was distributed to nurses working in all duty shifts (morning, evening and night) and who were providing direct care to patients (n=803). We excluded nurses with less than 12 months of seniority.

Study instrument

The study adapted an anonymous self-reported survey questionnaire. It was composed of 3 main sections: (a) demographic and workplace data (age, gender, marital status, educational levels, years of work experience, (b) questions related to WPV against nurses perpetrated by patients and/or parents or visitors within the preceding year: physical violence (PV) and verbal violence (VV), (c) mental health effects using the General Health Questionnaire (GHQ12) [5]. The reliability for the GHQ12 was determined by Cronbach´s alpha (α = 0.84).

Data collection

A pilot study adapted on 15 participants had been conducted in occupational health department to test the comprehensibility and usability of questions. The nurses answered the questionnaire in an average of 15-20 minutes and did not make any recommendations for changes. Before conducting the study, the researchers obtained approval from the directors of the two university hospitals of Sfax-Tunisia. Then, they went to all departments (except administrative and technical ones), distributed the questionnaire and invited nurses to participate. The questionnaires were answered individually, during working hours, in accordance with every nurse timetable, for causing minimum disturbance to the service, and with flexibility to fit working patterns.

Data analysis

Statistical analysis was performed using the Statistical Package for the Social Sciences (version 23.0). The results of quantitative variables were presented as mean ± standard deviation (SD). Qualitative variables were presented as percentages. Chi-square and Fisher´s test were performed for categorical variables in independent samples. In order to determine the independent factors associated with WPV, PV and VV, binary logistic regression analyse was used, after adjusting on the confounding variables. Adjusted odds' ratio (OR) with corresponding 95% confidence intervals were presented. To assess the impact of workplace violence on mental health, T-test of hypothesis testing was conducted to find any significant difference in mean GHQ-12 between those nurses who experienced workplace violence and those who did not. P-values lower than 0.05 were considered statistically significant.

Ethical considerations

To protect confidentiality, no identification information of participants was collected. Completion and return of the questionnaire was considered as consent to participate. There were not financial or non-financial incentives for survey participation. In each department, the head nurse was charged to supervise each participating nurse to collect the answered questionnaire in closed envelope.

Results     

A total of 276 participants returned the questionnaire for a response rate of 34.3%. Of these, 190 questionnaires were valid: 33 incompleted and 53 were excluded (less than one year of seniority). The age of participants ranged between 23 and 59 with a mean of 38.7±9.3 years, 66.3%were female. The majority held university degrees in nursing (70%) and had a mean work experience of 13.8±9.4 years. Of the nurses surveyed: 64.7% worked in medical department. The largest proportion (55.3%) had an alternate shift schedule (Table 1).

Workplace violence

The research found that 56.3% of the nurses were exposed to WPV from patients and/or parents or visitors. Of the participants violenced, 42.9% of the nurses had been subjected to physical violence and 93.4% of the nurses had been subjected to VV. Nurses stated that PV had come in the form of jostling in 39.1% of cases and of hitting in 36.9% of cases. VV against nurses was in the form of swearing in 53% of cases (Table 2).

Factors associated with WPV

A chi-square test showed that none of the variables tested were associated with WPV. Male nurses experienced more PV than female ones (OR=2.83; 95% CI=1.43-5.62). Working in psychiatry department and in pediatrics was significantly associated with PV (OR=4.76; 95% CI=1.66-13.63) and (OR=4.17; 95% CI=1.20-14.37) respectively. VV was more frequent in younger nurses (age ≤ 30) (OR=2.04; 95% CI=1.01-4.12). It is also more frequent in Bachelor´s degree nurses (OR=2.03; 95% CI=1.08-3.82). In contrast, married nurses, those working in intensive care unit and those working in morning shift were less subjected to VV (OR=0.36; 95% CI=0.15-0.87), (OR=0.27; 95% CI=0.08-0.87) and (OR=0.5; 95% CI=0.26-0.92) respectively (Table 3).

Logistic regression analysis of WPV

Table 4 presents the results of logistic regression analysis. After adjustment for confounding variables (age, gender, marital status, educational level, department and shifts), WPV was significantly associated with working in intensive care unit (OR=0.19; 95% CI=0.05-0.63). PV was associated with male gender (OR=0.21; 95% CI=0.09-0.48). Likewise, working in psychiatry department and in pediatrics increased the odds of experiencing PV (OR=8.64; 95% CI=2.22-33.5) and (OR=9.05; 95% CI=2.65-30.9) respectively. VV was significantly associated with working in intensive care unit (OR=0.24; 95% CI=0.007-0.82).

Impact of WPV on nurses´ mental health

The score of GHQ12 of participants ranged between 0 and 12 with a mean of 2.62±2.94. The comparison results showed that the mean score of GHQ-12 in nurses who experienced WPV and PV are higher than in those who did not (respectively p=0.045 and p=0.01) (Table 5).

Discussion     

This research was conducted to analyze violence experienced by nurses employed at two university hospitals in south Tunisia. The research found that 56.3% of nurses had been subjected to WPV the previous 12 months. This rate was higher than the prevalence shown in developed countries such as Italy (49.4%) [6] and Australia (36%) [7]. It was concluded that VV was more frequent than PV (93.9% versus 42.9%). Almost studies have shown that VV was higher than PV [8,9] because verbal abuse could not be controlled by any sort of security measures [8].

PV affected 42.9% of nurses which was higher than the rate shown in the study of Zampieron et al. (18.4%) [6]. However, VV reported in our study (93.9%) was similar to the same study (95.2%). This prevalence considerably exceeded that of both Palestine (40%) [10] and South Africa (45%) [11]. Coming out of a political and social revolution, violence has risen in Tunisian society in its most varied forms and become a daily mode of expression [12]. Male nurses were more often victims of physical violence than their female counterparts which was in accordance with a systematic review published by Edward et al. [13].

In our study, nurses working in intensive care units had experienced less WPV and VV than nurses in the other departments. These findings disagreed with those of previous studies noticing that this population shared the second rank of most frequently reported departments where WPV took place after emergency department [14,15]. This can be explained by the fact that violence is routinely underreported in this health sector. The intensive care workers considered WPV as an expected part of their job calling it “nursing silent epidemic” and even may be accepted it calmly [16]. Nurses working in pediatrics were more likely to have experienced PV. As shown in literature, pediatrics was known traditionally as high-risk for WPV [17]. The same applied to psychiatry nurses who were more likely to have experienced PV. In fact, previous studies have documented that 64.3-84.1% of psychiatry nurses were exposed to PV and are exposed to PV [6,18]. We found that WPV was related to poor mental health. Likewise, a Chinese study who found that the workplace violence had negative effects on the job performance and quality of life [19]. A systematic review concluded that violence is associated with negative consequences in healthcare such as increases in anxiety, depression levels and guilt with negative outcomes on well-being and quality of life of healthcare professionals [20].

To our knowledge, this study is the first to assess WPV among nurses in the Arab Maghreb Union, and to evaluate mental health using GHQ12 which is considered to be a high reliability and validity questionnaire. On the other hand, this study has several limitations. The questionnaire employed in the present study was based on retrospective data and self-reporting, which may have suffered from recall bias. Furthermore, the nurses' reluctance to participate is also a limitation of this study, as it is difficult to obtain information from unenthusiastic nurses. To solve this restriction, a larger-scale, multi-centric, comparative and prospective investigation could be conducted among various hospitals and health structures.

Conclusion     

While the findings of this study are generally not new, they do offer an illustration of the types and contexts of violence in the workplace towards nurses, including the type of nurse that is more likely to be targeted. More specifically, physical aggression was most frequent with male nurses, in psychiatry departments and pediatrics. Unlike literature, intensive care departments were protected from WPV. Nurses who were exposed to WPV often experienced a psychological impact post incident. Our study demonstrated that the WPV with nurses was associated to poor mental health. Determining the pattern and nature of WPV is the first step in developing suitable strategies to combat violence in the hospital workplace, and would support planning of appropriate intervention/prevention strategies. As hospital nurses still suffer from workplace violence, especially verbal abuse from patients/families, our results support the implementation of “safe hospital” policies and showing zero tolerance to workplace violence toward health care workers, including nurses, in order to retain them and to improve patient safety.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

Nada Kotti: writing and design methodology; Feriel Dhouib: writing and development of research aims; Amel Kchaou: data collection and application of statistical to analyse study data; Hana Charfi: data collection; Kaouthar Jmal Hammami: bibliographic research; Mohamed Larbi Masmoudi: activity planning and execution; Mounira Hajjaji: management and coordination responsibility. The authors have read and agreed to the final manuscript.

Tables     

Table 1: demographic and work characteristics of the participants

Table 2: types of workplace violence among nurses

Table 3: factors associated with WPV

Table 4: logistic regression analysis of workplace violence

Table 5: impact of WPV on nurses´ mental health

References