Prevalence and Consequences of Verbal Aggression among Bank Workers: A Survey into an Italian Banking Institution

: Robberies and other violent acts by strangers and intruders are common in banks, which can lead to post-traumatic stress disorder in employees. However, the literature indicates that more subtle and less measurable forms of violence, such as psychological violence, can also cause discomfort among employees. The aim of the present work is to investigate the prevalence of verbal aggression and its consequences among 311 Italian bank employees. A self-administered questionnaire was completed by 197 employees. The results showed a high prevalence of verbal aggression in the sample. It is important to investigate and ﬁnd the causes and effects of verbal aggression in order to understand the phenomenon in this workplace. Cases of verbal aggression need to be prevented and responded to appropriately. In addition, it should be noted that the effects of verbal aggression do not only affect the victim, but also the workplace, as productivity decreases, and the feeling of a safe environment is missing.


Introduction
It has long been known that banks are frequently exposed to robberies and other violent acts committed by strangers and intruders, which can cause post-traumatic stress disorder in workers (Fichera et al. 2015;Hansen et al. 2014). However, in recent years the literature has suggested that the deregulation of labor markets, emerging technologies, and new types of jobs have significantly reshaped working lives (Giorgi et al. 2017), and employees are frequently exposed to more subtle and less measurable forms of violence, such as psychological violence. Workplace violence is defined by the WHO as "the intentional use of power, including threats of physical violence, against another person or group that may result in harm to physical, mental, spiritual, moral, or social development. It includes name-calling, bullying, harassment, and threats" (Cooper and Swanson 2002, pp. 8-9). Workplace violence is often associated with repeated behavior that has negative consequences (emotional and physical) for the victim (Magnavita et al. 2019;Ricciardelli and Power 2020;Rudkjoebing et al. 2020). As Meyer and Kirsten (2014) pointed out, this form of violence is characterized by negative verbal and nonverbal aggression that can manifest in the workplace in the form of false accusations, criticism, unprofessional behavior, withholding information, excessive supervision, unrealistic expectations, and work overload. These behaviors usually take place in public to humiliate the victim. Whether negative behaviors are classified as verbal or nonverbal aggression also depends on the victim's perception. A main characteristic of verbal aggression is that it affects the victim's health (Hauge et al. 2010;Magnavita et al. 2021). It can originate from supervisors, colleagues, subordinates, and others (Keashly 2021). Aggression by superiors is related to a power imbalance characteristic of psychological violence, which refers to the victim's inability to fight back against his or her superior (Walsh and Ellis 2006). Individual perceptions of aggression also contribute to the experienced severity of an incident. Victims who perceive an incident as stressful appear to suffer more severe effects. In addition, Keashly (2021) found that perceived aggression and its severity also appeared to increase when coming from supervisors and peers.
The banking sector is an important component of the economy and is indispensable in modern societies. The European Banking Federation states that banks in the EU-28 employ about 2.7 million people, with 281,865 employees in Italy. As Njuki et al. (2013) noted, the introduction of ICT in banking and the increasing complexity of the products and services offered in ever-changing organizational structures have demanded more and more skills and competencies from employees. In addition, the new way of working in the world of banking and insurance exposes workers to customer complaints. As Van Greuning and Bratanovic (2020) pointed out, rapid innovation in financial markets and the internationalization of financial flows have changed the face of banking. The new practice is almost unrecognizable compared to banking as it was a few decades ago. To sell financial and insurance products, bank management encourages a more competitive behavior among colleagues (Mualla 2011;Seyed Javadin et al. 2020). Such factors may indirectly act as antecedents to violence: for example, management behavior could lead to a competitive and conflictual work environment (Salin 2003(Salin , 2005Tuckey et al. 2009;Djurkovic et al. 2021) and, in extreme cases, persecution. Kim and colleagues ) emphasized that banks are traditionally viewed as good work environments with relatively high annual salaries and good educational backgrounds and, therefore, are not a priority research topic. However, there are examples of bank employees committing suicide due to depression resulting from psychological distress, such as performance pressure. Thus, the impact of changes in the banking sector deserves special attention, as these changes have led to an increase in psychosocial disorders among employees, and the related work stress has reached critical levels in this sector (Giorgi et al. 2017). As suggested by Giga and Hoel (2003), work-related stress can be both a cause and an effect of violence. Longitudinal Italian studies have confirmed that workplace violence causes stress and that distressed workers are prone to violence (Magnavita 2013(Magnavita , 2014. Not surprisingly, Van den Bossche and colleagues (Van den Bossche et al. 2013), in their research to identify trends in the prevalence of workplace violence across Europe, found that the financial and corporate sectors in particular emerged as high-risk groups, confirming previous research (see for example Boyd 1995;Paoli and Merllié 2001). In a sample of 384 bank employees, Almeida (2003, cited in Verdasca 2011; see also Almeida and Diaz 2007) found that 56.3% of the respondents had experienced aggression during their working lives. Victims reported that they were not promoted by their department head, 47.1% reported that they were afraid to defend their rights, 76.4% reported that they were constantly given new tasks, and 4.7% reported that they were given inappropriate or unnecessary tasks. In the study conducted by Yilmaz and Uzuncarsili-Soydas (2006) on a sample of 200 Turkish bank employees, 15.9% of the respondents reported having been victims of aggression in the last six months. The most-experienced behaviors were being assigned tasks that were below the respondent's level of competence and withholding information that affected performance. Furthermore, Maciel et al.'s (2007) study of a sample of Brazilian bank employees (n = 2609) found that 7.9% of the respondents had been victims of violence at least once a week in the previous six months. Self-declared victims described the most common behaviors as being subjected to an unmanageable workload, work being harmful to mental or physical health, and receiving confusing or unclear instructions. Verdasca and Baillien (2021) found an incidence rate of 5.9% in their study of 561 bank employees. Interestingly, middle and upper managers and men were more at risk of being the target of workplace violence than women.
The purpose of this study is to assess the modern banking environment, in which verbal aggression (VA, hereafter) has become the most common and representative form of violence, and to determine the frequency of the phenomenon and its relationship to mental health. In the literature on violence in banks, events of type IV, involving physical violence or threats, are considered to cause PTSD, but VA is not. VA is usually brought forward by customers or colleagues, is much more frequent than robberies, and causes harm, especially in repeated episodes. As customer expectations may not be met by service organizations and their employees, this may trigger frustration and aggressive reactions (Yagil 2017). The aim of the present work is to obtain an up-to-date picture of the prevalence, consequences, and coping strategies of bank employees in a northern Italian banking institution. More specifically, the objectives are: a. to determine the prevalence of VA in a sample of an Italian bank employees; b. to evaluate the consequences of VA in terms of subjective emotional and behavioral reactions; and c. to evaluate the impact of VA on the presence of psychological symptoms of anxiety and depression.

Materials and Method
Experiences of workplace violence, anxiety, and depression were assessed with a questionnaire. The questionnaire included sociodemographic data (e.g., gender, age, and work experience) and a Violent Incident Form (VFI) (Arnetz 1998; Italian version by Magnavita and Bergamaschi 2008) to describe episodes of violence. In addition, anxiety was measured with the State-Trait Anxiety Inventory (STAI Y-1) and depression with the Beck Depression Inventory (BDI).
The VIF consisted of 11 questions (yes or no responses) describing a specific incident of workplace violence (physical or nonphysical) experienced by the worker. The first question addressed the frequency of aggression in life, the relationship to the perpetrator (e.g., a co-worker), the activity that preceded the episode of violence (e.g., a phone call), the nature of the assault (e.g., an insult), the victim's actions during the episode (e.g., "I asked for help"), subsequent responses (e.g., reporting the incident to the supervisor), and finally, the perceived causes of the incident (e.g., staff shortages). Victims were also asked to indicate the consequences of the physical and nonphysical workplace violence (e.g., no reaction, fear, anger, anxiety, humiliation, guilt, disappointment, helplessness, desire for revenge, feeling they made a mistake, or a change in work methods; in this study, Cronbach's alpha = 0.73). In this paper, we focus only on the verbal aggression subscale.
The STAI-Y1 (Spielberger 1983; Italian version by Pedrabissi and Santinello 1989) scale measured the state of anxiety, i.e., a person's current fear of an event characterized by subjective feelings of nervousness, tension, discomfort, and anxiety, accompanied by autonomic nervous system arousal. The scale included items such as "I feel tense" and "I feel secure". The scale included 20 items (response options from 1 = very severe to 4 = not at all) and total scores could range from 20 to 80, with varying degrees of severity: 40 to 50 was mild, 51 to 60 was moderate, over 60 was severe (Cronbach's alpha 0.85 and 0.84, respectively).
The BDI short form with 13 items (Beck and Beck 1972; Italian version by Sica and Ghisi 2007) is a scale for measuring depression symptoms. The response scale ranged from 0 to 3. An example of an item is "I am so sad or unhappy I can't stand it" (Cronbach's alpha = 0.88).
Carver's Brief COPE (Carver 1997; Italian version by Conti 1999) was used to evaluate the coping strategies used to manage verbal aggression. This was a self-administered questionnaire to assess 14 coping strategies: self-distraction, active coping, denial, substance use, emotional support, instrumental support, behavioral disengagement, venting, positive reinterpretation, planning, humor, acceptance, religion, and self-blame. Twenty-eight coping responses under stressful conditions were assessed, with scores ranging from 1 (I would not normally do this) to 4 (I would usually do this) (Cronbach's alpha = 0.71).

Procedure
The questionnaires were self-completed, and the participants received no compensation for their participation. The research project was approved by the Ethics Committee of the University of Turin (prot. N. 3173). Letters with details about the research project were sent to the managers of several local bank branches. Appointments were made with those who agreed to participate to explain the purpose of the research and to organize the completion of the questionnaire. An internal memo was sent to all employees informing them of the opportunity to participate in the survey. Copies of the questionnaire were distributed on dates arranged with the local bank branch managers. Additional copies were left for employees who were absent or off-duty that day. Employees who agreed to participate were asked to complete the questionnaire and place it in a box in their coffee kitchen (this was accessible only to employees; customers or third parties did not have access to it). The questionnaires were collected two weeks later.

Sample
Before conducting the survey, we calculated the sample size. Starting from the consideration that the total number of employees at the bank was 311 and not knowing a priori the percentage of people who had suffered violence, we hypothesized a prevalence of 50%. The minimum number of observations to obtain a 95% confidence level and a 5% margin of error was 173. The 193 responses we collected allowed enough leeway. However, we recalculated the minimum number of the sample considering the real frequency of reported verbal violence (26%). The correct calculation indicated a sample size of 152 cases. Our observations, therefore, exceeded the minimum sample size by 28.8%. The calculations were performed using Calculator.net (https://www.calculator.net/sample-size-calculator.html? type=1&cl=95&ci=5&pp=26&ps=311&x=123&y=18 (accessed on 15 June 2021)). The study sample consisted of 197 participants out of a total of 311 bank employees (participation rate 63.24%), of which the majority (N = 105; 53.8%) were from the banking sector, and 90 (46.2%) were from the insurance sector. The sociodemographic characteristics of the sample are shown in Table 1.

Statistical Analyses
Descriptive measures (percentages) were calculated for all the categorical variables (sociodemographic characteristics, consequences of VA, and subjective responses to VA). χ 2 -tests were used to examine differences in the prevalence of VA by gender, age group, industry, contract type, and occupational qualification. As a post hoc test, standardized Pearson residuals (SPRs) were calculated for each cell to determine which cell differences contributed to the results of the χ 2 test. SPRs whose absolute values were greater than 1.96 indicated that the number of cases in that cell was significantly greater than expected (in terms of over-or underrepresentation) if the null hypothesis was true, with a significance level of 0.05 (Agresti 2002). T-tests were used to examine the coping strategies of victims and nonvictims of VA. Finally, two linear regression analyses were conducted to understand the impact of experiences with VA in the past twelve months on anxiety and depression scores. Differences were considered statistically significant when p < 0.05.

Results
A total of 51 participants (26.2%) had experienced verbal aggression at work in the past 12 months. Table 2 shows the distribution between VA victims (VVA) and nonvictims (Not-VVA) differentiated by sociodemographic categories. Individuals aged 40 to 49 were most affected by the VA phenomenon, while individuals aged 50 to 59 appeared to be relatively protected. Bank employees were more likely to be affected by VA than insurance employees, and individuals with a permanent employment contract were also significantly more likely to report being victims of VA than individuals with a temporary employment contract.
Regarding personal perceptions of the direct consequences of VA, 62.5% of VVA reported feeling anger, 40.3% frustration, 33.3% humiliation, 33.3% helplessness, 31.9% disappointment, 19.4% worry, 15.3% concern, 11.1% fear, and 6.9% guilt. An amount of 11.1% reported feeling no consequences. As subjective reactions, VVA indicated that 47.9% felt a desire to change jobs, 42% felt personal inadequacy, 36.6% felt a desire for revenge, 29% were thinking about changing the way they work, 20% had a feeling of having done something wrong, and 15.5% were seriously thinking about changing jobs.
As shown in Table 3, there were significant differences in the use of coping strategies. VVA used significantly more maladaptive coping strategies (self-distraction, substance abuse, behavioral disengagement, and venting), except for a greater use of emotional support. Conversely, Not-VVA were more likely to use the coping strategy of adaptive planning. Anxiety scores ranged from 21 to 74, with a mean of 40.39 (ds = 11.1), whereas depression scores ranged from 0 to 29, with a mean of 5.02 (ds = 4.9). Two linear regression models were used to evaluate the influence of VA experienced in the previous twelve months on anxiety and depression scores (Table 4).
To assess linearity, two scatterplots of anxiety and depression as functions of being a victim of VA were created and overlaid with a regression line. A visual inspection of all four plots showed a linear relationship between the variables. The independence of each variable was confirmed with Durbin-Watson statistics of 2.091 (anxiety model) and 1.868 (depression model). Homoscedasticity and normality of the residuals were present. There were no outliers, so no subjects had to be removed from the analysis.

Discussion
More than a quarter of the observed bank employees experienced verbal aggression at work in the past 12 months, confirming earlier findings by Yilmaz and Uzuncarsili-Soydas (2006) and Almeida and Diaz (2007). This confirmed that the phenomenon needs to be studied in order to prevent it. In this study, it was found that the attacks were mainly aimed at males, although the difference with females was not significant, according to the observations showing that men were more likely to be victims of verbal aggression than women (see Guay et al. 2014 Salin (2003). They each found that these workers had low status in their careers in terms of pay and job security and tended to achieve better roles. This created a power imbalance that fostered aggression. The feeling of relational powerlessness is usually one of the main causes of victimization (Roscigno et al. 2009;Cowan et al. 2021). A majority of the victims had a permanent position, and they worked in the insurance sector. As Walter (2004) pointed out, mergers in the banking and insurance sectors have led to restructuring risks, losing cultural values focused on the proper management of human resources. In addition, financial crises can be a source of stress, so one possible consequence is aggression among bank employees (Salampasis et al. 2015). Workers were shown to suffer particularly from the deterioration of business performance, which could determine structural adjustment and performance constraints and improve employment anxiety (Lee et al. 2022). This is especially true in the insurance sector. Employees working in call centers provide service and support remotely via telephone and Internet technology, reducing the need for extensive branch networks and direct customer contact (Giga and Hoel 2003); therefore, they might experience criticism from customers with great personal distress.
Interestingly, the majority of VVA stated that they were inadequate (personal inadequacy or feeling they were doing something wrong) and that they had been working inadequately (thinking about changing the way they work). This feeling characterizes victims of all forms of violence and is linked to the perception of low self-esteem (O' Moore and Kirkham 2001;Valera-Pozo et al. 2021). Employees with low self-esteem may be perceived as vulnerable to VA due to their lack of self-confidence, which may make them reluctant to retaliate. However, as Samnani and Singh (2012) pointed out, the causal direction of relationships between aggressive behavior and self-esteem has been questioned, as low self-esteem can develop and be reinforced when aggression persists over time. In a study by Rossiter and Sochos (2018), perceptions of social support, particularly support from the work environment, were found to contribute to workers staying at work despite a perceived negative evaluation of their own job skills, which was a consequence of low self-esteem. Participants who reported being VVA experienced feelings of anger, frustration, disappointment, and helplessness. The latter indicates that they did not feel adequately supported in their workplace. The results of the Nielsen et al. (2020) study suggested that social support, in particular support from supervisors, could reduce the negative effects of violence on individuals' health and ability to work. Companies should, therefore, incorporate social support into policies to address violence. The presence of social support in an organization has been identified as an important organizational factor that can enhance employee wellbeing and reduce stress and strain. Because social support affects stigma in seeking help, perceived organizational support increases the possibility of contacting counseling services for individuals, both victims and nonvictims (Acquadro Maran et al. 2022).
The search for support in the workplace was indicated in the strategies used by VVA to cope with the phenomenon: they were more prone to use emotional support than the Not-VVA. Emotional support is considered an adaptive strategy. Support from friends or family members is associated with higher levels of personal achievement. In addition, VA at work was associated with high turnover intention, and social support played a buffering or mediating role in the effects of VA on work-related outcomes (Rudkjoebing et al. 2020). However, Jeong and Kim's (2018) study showed that emotion-oriented strategy was one of the factors related to the intention to leave the workplace. Seeking peer support could be a strategy to express negative feelings and obtain support, but it could not change the negative context in which the phenomenon occurred because they did not know how to intervene (Johnson 2019). This creates a climate of frustration that affects the entire organization: previous research has shown that witnesses to workplace aggression perceived more stress and had more mental health problems than nonwitnesses (Sprigg et al. 2019;Acquadro Maran et al. 2021). Furthermore, VVA were more prone than Not-VVA to use maladaptive coping strategies, such as self-distraction, substance abuse, behavioral disengagement, and venting. These strategies are common among victims of all types of workplace violence and are symptoms of a lack of coping skills. In addition, long-term maladaptive coping led to a lower adaptive capacity that allowed workers to experience high-stress situations with long-term physical and psychological consequences (Vrablik et al. 2020). Regarding psychological consequences, findings have shown that the experience of VA predicts both anxiety and depression symptoms. As shown by the investigation of Rodríguez-Muñoz et al. (2020), as exposure to aggression in the workplace increased, so too did symptoms of anxiety and depression. A systematic review by Boudrias et al. (2021) aimed to analyze the literature on the long-term consequences of workplace aggression and the mechanisms involved. The results showed that the phenomenon was a precursor to psychological impairment in victims and that this relationship persisted over time.
Research has suggested that depression symptoms persist for up to four or five years in employees exposed to aggression (Bonde et al. 2016;Einarsen and Nielsen 2015) and that the association between exposure to the phenomenon and depression is particularly strong (Einarsen and Nielsen 2015).
Overall, the findings from this investigation suggested the need to prevent VA in the banking sector. One of the most important problems in dealing with VA is the lack of information on the part of organizations. This lack can affect both those who need to help victims, such as managers, who very often have difficulty providing adequate assistance because they are unfamiliar with the phenomenon, and the victims themselves, who are not trained to deal with VA. Since VA in organizations can be a process in constant pejorative evolution (Acquadro Maran and Begotti 2020), so too must effective prevention start at all organizational levels, in general, ensuring that organization policies against inappropriate workplace behavior are clearly communicated. In this way, all employees know there is zero tolerance for VA and other types of workplace violence. A lack of sanctions for those who behave inappropriately can lead to the assumption that it is legitimate. Instead, a clear and well-defined company policy helps ensure that VA is recognized as such, especially by the victims themselves, who may view this behavior as "annoying but tolerable". At the management level, a useful training method could be dealing with high-conflict situations and contexts. At the human resources level, which is the basic reference point for VA cases, it might be useful, for example, to conduct a scheduled survey of the perceived organizational climate. In addition, it might be useful to provide support, such as a counseling service, for those employees who need assistance. During the interview, the professional could offer suggestions for coping with stressful situations and explain various coping strategies and their consequences at both individual and group levels. For employees who are perceived as VVA, it might be useful to help them restore their self-esteem through specific techniques, such as "conflict training" (Baixauli et al. 2020). The conflict training could include courses on verbal self-defense and the M-group. Courses on verbal self-defense are personal training courses that aim to teach the rules and basic strategies of defense against verbal aggression in order to block and undo it (Wirth et al. 2021). An M-group lasts three days and is structured like a classic T-group, that is, it takes place in an isolated place where both the participants and the instructors are in exclusive and constant contact without any external influences. The participants deal with a simulation in which the leader provides a background structure to recreate a typical work situation. The goal is to recreate a work conflict so that the participants understand how it works and can find a possible solution (Pastore 2006;Braverman 1998). In addition, in cases that require more targeted intervention, it may be appropriate to recommend a counseling service that can help the victim and the witnesses. In this case, it is important to create a climate that eliminates possible stigma, victim-blaming, and other barriers to the use of mental health professionals (Rodrigues et al. 2021).
This study has several limits. First, the cross-sectional design made it impossible to infer on causality. However, the suggested relationships between violence and the reported effects seemed plausible. Another limitation is that we analyzed only one company, which limits the generalizability of the results. The study was conducted in a single bank that employed 311 workers. Of these, more than 60% participated. Further research could include a larger sample with other banking institutions. This could allow a more-detailed examination of VA and identification of the individual and organizational factors that might prevent violent behavior. Thus, because the sample was too small, we could not disaggregate the data by the sexes of the victims and the perpetrators. However, as noted in previous research (Maran et al. 2019;Salin 2021;Otterbach et al. 2021), gender may play an important role in the perception of VA and the experience of consequences and coping strategies. For future studies, it would be useful to consider the role of the genders of the two actors involved in VA and to analyze whether and how it affects the relationship between them. In addition, it is difficult to identify oneself as a victim of VA in the workplace. One of the reasons for this difficulty is the fear of being mistaken for a victim of persecution (Al-Atwi et al. 2021). Furthermore, victims may deny or minimize the incidents. Denial and minimization are coping strategies that allow one to mitigate unpleasant feelings and rationalize the other person's behavior; indeed, most VVAs considered themselves inadequate. These strategies can help the individual ignore the situation and potential dangers and go about his or her activities and usual routine. As described earlier, sharing the experience with other colleagues, friends, or relatives can be helpful in getting the support needed. Further research could analyze possible organizational barriers to seeking organizational and social support and reporting VA. Another limitation concerns possible participation bias: there could be a bias related to socially desirable responding, i.e., the tendency to answer a questionnaire while projecting a positive image of oneself. We did not use a social desirability scale (Van de Mortel 2008) in conjunction with the other scales used in this study. To improve the validity of questionnaire-based research, it could be included in future studies. Finally, as described by Van Greuning and Bratanovic (2020), the environmental context in banks has changed over the past two decades. Future studies could examine from a qualitative perspective how these changes have altered work practices. Employees with years of experience could describe how the environmental context has changed, the significance of these changes in terms of organizational and emotional context, and the different types of violent behavior they have experienced during their work experience. Despite this limitation, the results of this study illustrated that these are employees that organizations need to protect, and therefore, research on these employees should continue.

Conclusions
Overall, the results showed the high prevalence of VA in a sample of employees of a bank in Italy. It was important to investigate and find out the causes and effects of VA in order to understand the phenomenon in this workplace. Cases of VA need to be prevented and responded to appropriately. In addition, it is important to note that the impact of VA not only affects the victim, but the workplace is also impacted by a decrease in productivity and a lack of perception of a safe environment. One of the best ways to eliminate VA is to stop it before it starts. In other words, organization policies that incorporate VA can deter the phenomenon; the goal is to reduce work-related stress by fostering a strong sense of trust within the organization, providing certain norms, and instituting a "zero tolerance" policy toward violent behavior. As Di Fabio et al. (2016) noted, it is particularly important to create a climate of mutual support and to foster an organizational culture that enables the adoption of behaviors focused on respect and civility. For VVA, banks should provide specific therapy to reduce the consequences of victimization (e.g., anxiety and depressive symptoms, rumination, fatigue, exhaustion, and depersonalization). This support should aim to restore confidence in the victim's own ability to cope with stressful events, such as managing interpersonal conflicts at work. However, the best strategy to prevent verbal violence is undoubtedly to improve relations with the public. The training of bankers should aim to enhance characteristics of empathy and extroversion, which can improve relationships with customers. Employees must be educated to honestly report to customers not only the benefits, but also the risks of investments. The employee must position himself not as a seller of financial instruments, but as an intermediary between the customer and the world of finance, clarifying its role and the limits of the options proposed. In this way, the customer is able to correctly perceive the bank's action as a service and not as a trap, and they are not led to blame employees for the effects of economic crises.
In summary, organizational measures are needed to implement mandatory workplace health programs and workplace health promotion programs. These solutions can be useful for the protection and promotion of workers' mental wellbeing and for the emotional support of VVA. New ways of training bankers are useful to reduce the frequency of verbal aggression by customers. Further longitudinal studies are needed to understand the mechanisms linking VA and mental health. Further research should also examine the prevalence of the phenomenon, the type of behavior exhibited, gender differences in victims, and the relationship between perpetrator and victim.

Institutional Review Board Statement:
The study was conducted in accordance with the Declaration of Helsinki, and approved by the Ethics Committee of Università di Torino (protocol code 3173, 27 January 2016).

Informed Consent Statement:
Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The datasets generated for this study are available on request to the corresponding author.