Experiencing violence in a psychiatric setting: Generalized hypervigilance and the influence of caring in the fear experienced

Abstract

BACKGROUND: Exposure to violence in the mental health sector both affects employees and has implications for the quality of care provided.

OBJECTIVE: This phenomenological study aims to describe and understand the ways in which acts of aggression from a patient might affect workers in a psychiatric institute, their relationships with the patients and the services offered.

METHODS: Two semi-structured interviews were conducted with each of the 15 participants from various professions within a psychiatric hospital.

RESULTS: Our analysis reveals four themes: hypervigilance, caring, specific fear toward the aggressor and generalized fear of all patients. A state of hypervigilance is found among all participants. An emphasis on caring is present among the majority and unfolds as a continuum, ranging from being highly caring to showing little or no caring. A feeling of fear is expressed and is influenced by the participant’s place on the caring continuum. Caring workers developed a specific fear of their aggressor, whereas those showing little or no caring developed a generalized fear of all patients. Following a violent event, caring participants maintained this outlook, whereas those demonstrating little to no caring were more inclined to disinvest from all patients.

CONCLUSIONS: Hypervigilance and fear caused by experiences of violence impact the quality of care provided. Considerable interest should thus be paid to caring, which can influence fear and its effects.

Keywords

Agression workplace quality of care phenomenology

1 Introduction

In recent years, workplace violence has received increasing attention by employers and researchers and its many negative consequences have been revealed. The healthcare sector involves greater risk of workplace violence [1], especially in hospital emergency rooms and psychiatric facilities [2]. Episodes of violence not only affect the health of employees, but could also impact on the interactions that workers have with patients, as well as on the quality of care provided [3]. Certain behaviors, such as fear and avoidance of patients [4], have been observed among healthcare workers who were victims of violence. This article focuses on the violence perpetrated by patients in a psychiatric hospital and its impact on the interactions between healthcare workers and patients.

1.1 Workplace violence

Workplace violence includes physical assault, threats of assault and psychological aggression [5]. Dupré and Barling [6] define workplace violence as “any behavior intended to harm an individual in an organization”. Serious violent acts (SVAs) are defined as behaviors leading to physical injuries as well as threats of such acts [7]. For the purposes of the present study, SVAs include physical abuse, theft with violence, armed robbery, assault, sexual touching and threats of death or injury.

1.1.1 Identity of the aggressor

Acts of aggression in the workplace are usually committed by someone known to the victim and are most often committed by a member of the public or a client/patient that comes into contact with the victim [8], especially in psychiatric settings [9]. In the present study, we focus on acts committed by patients in a psychiatric hospital, which corresponds to “Type II workplace violence” [10]. Previous studies have investigated the impact of these violent acts on the targeted healthcare worker.

1.1.2 Consequences of type II violence

Patient violence has negative effects on the healthcare workers targeted by this aggression [3, 11]. The consequences may be physical, psychological, financial, or represent an overall impairment of functioning. Physical consequences include the visible injuries inflicted on the body of an employee [12]. Psychological injuries can include irritability [13, 14], anxiety [15], depression [16], post-traumatic stress disorder [4], anger and shame [15], fear and mistrust [16], guilt [17] and helplessness [14, 17]. Workplace violence can also have financial consequences for institutions, including replacement costs during absences, sick leave and compensation payments [18]. Finally, overall impairment of functioning includes decreased motivation [16], desire to quit one’s job [19], questioning one’s professional skills [19], diminished sense of security [20], decreased job satisfaction [21], absenteeism [22] and a decline in workplace performance [16]. This set of consequences is closely linked to the quality of services that are provided [4 , 23].

1.2 Effects of type II violence on the quality of care

Exposure to violence in the health sector may affect interactions with patients and the quality of provided care [3, 24]. To date, several studies have addressed the negative impact of workplace violence on employee productivity, as well as on their ability to provide appropriate care [4]. Moreover, workplace violence has evoked identifiable responses in healthcare workers toward their patients, including fear and avoidance [4, 25], diminished propensity to meet their demands [26], insensitivity and even lack of empathy, and passivity [15]. Furthermore, Warelow, Edward and Vinek [27] suggest that trauma, verbal attacks or anxiety-provoking incidents can lead nurses to an unconscious decision to move toward an uncaring part of the caring continuum.

1.3 Objectives of the study

Our aims are to describe and better understand how victimization affects healthcare workers, their relationships with patients and the quality of services they provide in a psychiatric setting. Specifically, this study aims to explore factors that influence the impact of violence on these workers. This exploration may inform the development of strategies to address the aftermath of workplace violence and contribute to violence prevention initiatives. We will attempt to answer the following questions: What is the experience of healthcare workers who have been subjected to violence by patients? How does this experience alter both the quality of relationships with patients and the quality of care provided?

2 Methodology

The use of a qualitative and descriptive phenomenological approach allows for a different perspective on these issues. By immersing ourselves in the unique world of each participant, we can better understand participants’ perceptions and interpretations of the violence they have experienced. This approach is more appropriate for identifying what constitutes this phenomenon (rather than what causes it) and for acquiring a more comprehensive perception of (rather than control over) the victim’s own understanding of their experiences [28]. In addition, this approach will describe the participants’ perceptions of how the SVA has altered the services they offer and their interactions with patients. The use of the phenomenological approach provides access to extensive content in order to explore the ways in which the practice of healthcare workers is altered by this type of experience.

2.1 Participants recruitment

Participants were recruited from a psychiatric hospital in Montreal, Canada, in collaboration with the Service de santé et sécurité au travail (SSST), which is a health prevention service present within the hospital. This hospital was chosen for the study because it is proactive in the research, prevention and treatment of SVAs in a psychiatric setting, and was willing to pursue any conclusion or recommendation that would follow. The hospital is comprised of about 2,000 employees working directly with the patients. Between April 2014 and March 2015, 237 events of physical aggression were reported and 10 of these lead to work leave.

The researchers produced and distributed an information letter to eligible employees after meeting with members of the SSST. Authorization and consent forms were completed by employees interested in the study. The “snowball technique” of recruitment was used in conversations with participants, both over the phone and in person. This study was approved by the ethics committees of both the hospital and the University of Montreal.

2.2 Participants

Our qualitative approach allows for the recruitment of a non-probabilistic sample, but does not specify a certain number of required participants [29]. Benner [30] suggests conducting interviews with as many participants as needed until empirical saturation of the data is reached, i.e. when the only new information provided by participants is anecdotal. Although we planned to recruit approximately 20 participants between November 2012 and May 2013, this number remained flexible based on empirical saturation following the analysis of verbatim transcripts. The recruitment process ended once saturation was reached, thus generating a final sample of 15 participants. Interviews were conducted between January and May 2013. In addition to the selected participants, telephone contact was made with 11 other employees of the psychiatric hospital. Of these, five did not correspond to the selection criteria (e.g., current work leave, or no work leave following the SVA), five refused to participate for several reasons (e.g., lack of time, fear of reliving the SVA), and one withdrew his participation from theproject.

The final sample consisted of 15 participants. The inclusion criteria were: (a) being employed at the psychiatric hospital where the research was held; (b) having experienced a SVA in the past decade followed by work leave or absence and (c) having been back at work for a period of at least two weeks. The time period of 10 years was chosen in order to be able to evaluate the long-term impact of the SVA while assuring relatively good recall of the event. Including participants that experienced work leave or absence allows us to reach the individuals for which the events had severe consequences. Exclusion criteria were: (a) having a degenerative neurological disease; (b) having an organic brain disorder; (c) having intellectual disabilities; (d) having experienced past or present psychotic episodes; (e) having a diagnosis of major depression or bipolar disorder; (f) suffering from substance dependency or (g) having current suicidal thoughts. These exclusion criteria eliminate certain variables other than the SVA that could impair the participant’sfunctioning.

The sample included eleven women and four men, of an average age of 42, ranging between 28 and 62 years old. The study was open to all types of healthcare workers, and our final sample represented seven orderlies, five nurses, two educators and one auxiliary nurse. On average, participants had 18 years of experience in the healthcare sector, ranging from 4 to 36 years. Participants worked day shifts, with the exception of one who worked evenings and another who worked nights. Thirteen participants were employed full-time and two part-time. In terms of their experiences of SVA, three participants reported one incident, while the rest reported multiple events. Five participants had been victimized twice, four participants three times, and two participants eleven times. One participant reported experiencing approximately two SVAs every week. Twelve participants reported physical assaults, one had been the target of threats, another was subjected to sexual touching and another helped another worker who was being assaulted. Thirteen of these events occurred in inpatients settings and two while providing services within outpatient settings.

2.3 Data collection

Based on an interview guide, semi-structured interviews were conducted to identify themes related to their interactions with patients, prior and after their SVA. Each participant was interviewed twice to alleviate the content. The interview design served to differentiate between descriptions prior to and after the SVA and to validate the researcher’s understanding. The time lapse between the SVA and the first interview ranged between 8 and 57 months, with an average of 21 months. Each interview lasted between 60 and 90 minutes, and was recorded and later transcribed. All interviews were conducted by the principal author and took place between January and May 2013. During the first interview, participants were asked to complete a brief survey of sociodemographic information (e.g., age, sex, religion, marital status). Next, a series of open-ended questions were asked to encourage participant reflection on the kinds of interactions they had with patients prior to the SVA. The second meeting provided the researcher with an opportunity to validate her understanding of the first meeting by reading a summary of the latter to the participant. Appropriate changes were made. Participants were then asked to describe the SVA experienced as well as examples of interactions with patients following theincident.

2.4 Analysis

Transcripts were first analyzed using the “Empirical Phenomenological Psychological” method [31]. This analysis was conducted in five steps: (a) reading and re-reading transcripts to get an overall impression of the data; (b) division of the data into units of meaning; (c) identification and interpretation of themes; (d) summary of themes for each participant (“horizontal analysis”) and (e) overview of the general themes and sub-themes emerging across participants (“vertical analysis”). The final step allowed us to identify recurring themes addressed by the participants, and to observe the extent to which similarities and dissimilarities emerged between the individual experiences. NVivo 10 was used in our systematic analysis of the data. The transcripts were inserted in the software, which facilitated the division of data by units of meaning and the ulterior combination of those units to form the main themes. Note that all names have been changed to ensureanonymity.

3 Results

3.1 Hypervigilance

All participants revealed that, following their experience of a SVA, they found themselves to be in a persistent state of hypervigilance. This state involves an acute and heightened level of responsiveness to any element that might be potentially related to the traumatic event. According to participants, “caution” becomes necessary in a workplace now considered to be “stressful”, “dangerous” and “unpredictable.” Some even reported that simply witnessing violence similarly created a more vigilant attitude. Alex (orderly, 46) had been both a witness to––and victim of––various SVAs and expressed a constant state of caution as a result of these experiences:

We can’t imagine what it might be, the danger of working in a psychiatric unit. We’re always walking on eggshells… And the worries are always there. It becomes subconscious, but it’s always there.

This state of hypervigilance can emerge both in the workplace and in the personal lives of workers. This vigilance seems to be reactivated by stimuli that are somehow related to the context in which the violence took place. When the participant observed the patient becoming gradually disruptive— where the aggression could be anticipated— his or her sense of hypervigilance seems to be limited to the workplace. This state is extended to other areas of a worker’s life when the attack took place suddenly and without warning (e.g., when a participant was attacked from behind). The following excerpts illustrate these different contexts and their implications. The first describes the aggression experienced by Carla (orderly, 52), who observed a patient becoming disruptive before the assault. The second describes the context of aggression experienced by Sarah (educator, 53), who was attacked frombehind.

He took his plate and then threw it at the window of the dining room. So that he would stop pulling the trays, I removed them. Then he got up, but really fast. He took me by the wrists, and then assaulted me the same way he did with the evening crew. He did exactly the same thing. He was kicking my legs and then he tried to aim at my crotch (Carla, orderly, 52).

I was doing the rounds and I noticed that she was not in her room. Just a little further down, she had hidden herself behind a door. I hadn’t seen her. It was then that the assault occurred. I had my back to her, but I could not know that she was hidden behind the door. So when I felt myself being grabbed, I turned and tried to free myself (Sarah, educator, 53).

When the attack occurred unpredictably, hypervigilance seems to be extended to the participant’s everyday life. Sarah explains how she changed her habits to decrease the risk of violence. Her state of hypervigilance is sometimes noticed byothers:

Suddenly, I had become like an elderly person who is afraid of getting her handbag stolen, who is afraid of being attacked at the ATM, who is afraid of… everything. It had been two months… I went to a pharmacy and there was a lady who was following me really closely behind. It was like she was entering my bubble, you know, with her cart. I kept checking for my purse and she told me, “it seems like you don’t trust people.”

Sophie (educator, 29) also shares how this state of vigilance altered her daily life and specifically expresses not wanting to be unpredictably attacked again.

You know, there are things in my work, even in my daily life… like now, I lock my car doors when I’m in my car. I did not do that before. I don’t feel like being taken by surprise again. Let’s put it that way.

Following the SVA, two participants described their hypervigilance as a “survival instinct”, something necessary to “ensure survival.” This general state of insecurity creates a significant need to find ways of overcoming it. Participants mentioned that they were seeking changes at various levels to ensure survival: (a) in the work environment, measures might be taken to increase the number of security officers and surveillance cameras or to decrease the number of patients in a unit; (b) at the clinical level, treatment plans or intervention techniques might be reviewed, such as being more “polite” or less confrontational with aggressive patients; (c) at the personal level, participants might seek ways to empower themselves (e.g., through learning martial arts).

3.2 Caring

Caring implies a compassionate and authentic attitude toward the patient. Placing patients at the heart of his or her interventions, the caring participant develops a relationship of trust and acts as an agent of change. In this study, caring is perceived as a continuum ranging from being highly caring to little or not caring.

Caring emerges in participant accounts within three dimensions. These are patient-centered care, dynamic and interactive relationships, and the importance of trust. The interviews suggest that the presence or absence of one or more of these dimensions can actually influence the ways in which fear becomes manifest (i.e., specific or generalized).

Patient-centered care is defined by the fact that the patient is the primary concern of the healthcare worker. From this perspective, tasks and interventions are discussed as being patient-focused, specific needs of the patients are kept in mind, and there is an expressed wish to “take care” of them, to “bring positivity”, and to “work with them.” In these ways, participants who have a caring disposition constantly describe the patient as being at the heart of their interventions. Marie (nurse, 55) mentions that the extent to which she felt satisfaction with her day was based on the happiness of her patients, the assistance she had provided to them and the contact she had with them:

I liked the help I could give them, the few little things that gave them pleasure. You could see in their eyes that they were happy with their day, with the time we could give them.

Participants also sought out and appreciated dynamic and interactive relationships with patients. They saw themselves as agents of change in the patient’s life, contributing to his or her progression and wellness. Indeed, they found it motivating to establish “goals” and to educate, stimulate, guide, accompany and advise the patient over the course of their progress. Participants expressed feeling happy when they saw an improvement in a patient’s mental state or level of rehabilitation. Making a difference and contributing to the patient’s progress, was described as gratifying for the participant. As Isabelle (orderly, 33) notes:

You know, it is certainly gratifying when he arrives… then he calls you all kinds of names because he is completely out of it… he is completely delusional. Then, you know, we continue to do our job, we continue to provide the best service and after a few weeks, a few days, he’s better. I think that’s rewarding, you know, to see what we do… We don’t do it for nothing. We are contributing to making this person better.

Caring workers see this interactive and dynamic relationship as mutually beneficial and stress the importance of recognition. They are eager to assist patients and to give themselves completely, and they are constantly looking for ways to “provide assistance” and “to lend a hand.” In return, they appreciate recognition, which feels gratifying and satisfying to them. Lucie (auxiliary nurse, 32) confirms that receiving recognition validates and encourages her daily efforts.

They always end up giving back what you give them. You know, when they’re happy to see you, they always give back what you give them. It’s like that. When you’re smiling more, when you’re having more fun with them, they’ll tell you. And, you know, they give it back. They always end up giving it back. I think it’s good because I tell myself that I did not make all these efforts for nothing. Like, at least, you know, they are aware of it.

Finally, in the course of developing dynamic and interactive relationships with patients, caring participants prioritize the establishment of an authentic and trusting relationship with the patient. This gives the patient a sense of confidence that he or she can count on— and confide in— the worker, which, in turn, facilitates the cooperation of the patient even during the most difficult situations. Trust also gives the worker a sense of being a “useful” and capable “resource person”. Bianca (nurse, 28) illustrates this point: “I liked being in contact with patients, I liked trying to understand them, to help. You know, there are patients that you give more to, then you’re proud of what you have done, to have worked, for creating such a bond. This isn’t always easy with psychotic people.”

3.3 Specific fear and modified perceptions toward the aggressor

Interviews revealed that the effects of SVA on the relationship between the worker and the patient differ depending on where the participants are situated on the caring continuum. Among the most caring workers, a specific fear was felt toward the violent patient. In this case, participants hope for the aggressor to be absent from their unit when they return to work, fear seeing or being around him or her, and fear being the target of retaliation.

In the following excerpt, Sarah (educator, 53) describes the hindered relationship with the aggressive patient. Simultaneously, her perception of the patient’s prognosis has also changed; she now believes that there is no possibility of the latter’s improvement and describes his cognitive level of understanding as limited or nonexistent. A similarly altered perception of the aggressor emerged in the majority of participants’ accounts. Participants stated that there was “nothing to be done” with the aggressive patient and that he or she would “never change.” This perceived irreversibility may explain why we observe a withdrawal of aid and treatment for this specific patient.

It is his presence that was difficult. It was his presence. You know, I think when an event like that happens to us, the relationship is broken between the patient and the worker. Everything was broken for me. So I was doing the tasks that were assigned to me for this patient. This client wasn’t having psychotherapy, because it’s a person who is just action-reaction. If she doesn’t have what she wants, she gets frustrated and reacts. The level of understanding is: I don’t have what I want, so I attack.

For caring participants, their fear is specifically directed toward the aggressor, so they maintain a caring outlook with the rest of their patients. Lucie (auxiliary nurse, 32) stated that, outside of the contact she has had with her aggressor, she has maintained the same degree of appreciation for her work: “I still like the same things I liked before. I like taking care of my patients, that, that has not changed. It’s only toward this specific patient that it’s changed.”

Genevieve (orderly, 43) expressed that she has actually become more invested in patient activities since the assault. This heightened involvement is expressed in a greater desire to share her opinions with colleagues on matters pertaining to patients, such as their mental state, their individual difficulties or better ways of interacting with them.

Before, I would never take part in committees. Now, I invest myself much more. Before, I would let the others speak. Now, I give my point of view. Before, it was really more… I chatted with the patient, I was with the patient. Now, in addition to talking with the patient… Well, I know what to say, because I know more about his record. As soon as he arrives, they do a briefing right away… well, he has this, he arrived for that. So we know right away what to expect.

These polarized tendencies— of abandoning aid to the aggressor versus developing a deeper general investment in other patients— might possibly serve as a means for preventing future violence for caring workers.

3.4 Generalized fear and modified perceptions toward all patients

Among those participants who were closer to the less caring pole of the caring continuum, a sense of generalized fear emerged in relationships with all patients following the SVA. This sense of fear surpasses the state of hypervigilance described earlier. The fears and apprehensions are described as constantly present, leading the worker to question his or her ability to do the job— especially with agitated patients. Nathalie (nurse, 35) recounted that she had become different after the attack; overwhelmed by this sense of fear, she did not feel able to carry out her work properly upon her return.

I was completely different from before. I now had fear… I was really scared. Fear of not being able to do my job in the same way according to my skills and my judgment, and also not being able to intervene with agitated patients.

In connection with this feeling of generalized fear, participants developed altered perceptions of their patients. Nathalie (nurse, 35) expressed a lack of empathy toward patients and a loss of interest: “I have less sympathy toward patients. To be honest, I’m a little disillusioned. I don’t know how to say it.” Isabelle (orderly, 33) developed a similar modified perception of patients. In fact, before the assault, she viewed patients as individuals who had the ability to improve and develop through treatment. Following the assault, she began to describe patients as “limited people” of whom she had to be cautious and keep at a distance: “Patients will always be patients. They are limited physically or mentally. If he has a cane, well, you walk at that distance from him because you never know when you’re going to be hit.”

As Isabelle’s case demonstrates, participants with a generalized fear of all patients were apt to distance themselves and to disengage from their work. This detachment is as much behavioral (e.g., no longer working overtime) as it is emotional (e.g., disinvesting from treatment of patients). Nathalie (nurse, 35) demonstrates this point:

I am much less invested in general. I just do my job, really. I do the work, but I don’t have… Before, I really took it to heart. It was… I was right in there, passionate. Not anymore. Not at all. I just do the job. I provide care and I sometimes even tell myself, “Natalie, you’ve got to…” you know, I don’t give bad care, but I say… I have to push myself. I do what I have to do but not much more.

After the assault, some participants even changed work assignments in the hospital to perform less risky (and sometimes less interesting) tasks. This was the route taken by Isabelle (orderly, 33), who decided to avoid direct contact with patients as much as possible since the SVA and took on a cleaning job. It should be noted that before the assault, Isabelle appreciated contact with patients, as well as her role in encouraging and supporting them to make a difference in their lives.

Of course, everyone knows that cleaning is cleaning. It’s less interesting. Although [the tasks are] harmless. I don’t have the same responsibilities. If someone is sleeping and he doesn’t want me to go into his room, fine, I go to the next and it ends there. You know, I don’t need to insist. It’s not necessary… It’s not something that has big risks.

Our analysis suggests that, for the participants situated on the less caring pole of the caring continuum, generalized fear of all patients can develop as a result of workplace violence. This carries with it a tendency to withdraw and disengage from healthcarework.

4 Discussion

In this study, participants’ perceptions of their SVA were analyzed in relation to four themes. First, we found that all participants reported a state of hypervigilance, resembling an enduring habitual reaction that seemed to persist after the violent episode. In addition, a sense of fear emerged in the accounts of some participants. Findings suggest that this fear may be influenced, in terms of its generalization or specificity, by the position of the worker on the caring continuum.

4.1 Hypervigilance

Studies examining violence in healthcare settings demonstrate that patient aggression negatively affects the targeted workers. Much of the research addresses the negative emotional impact, without distinction between hypervigilance and fear. For instance, McKenna and colleagues [32] reported that 25% of nurses experienced confounded feelings of fear, anxiety, hypervigilance and distrust of patients following an assault by a patient in the first year of practice.

In regards to hypervigilance, which is sometimes associated with distrust [33], the results of previous studies are mixed. Crilly and colleagues [34] found that 44% of nurses working in emergency rooms became more cautious in their relationships with patients, and 50% of the healthcare workers in Arnetz and Arnetz’s study [24] expressed increased vigilance. A few studies found lower levels of vigilance; one revealed that only 4.3% of the sample were hypervigilant with patients [16]. Closer to our results, Cheung, Bessell and Ellis [35] found that the majority of nurses did not feel safe in their work environment and remained hypervigilant following an assault by a patient. Moreover, our findings are in line with Kindy, Petersen and Parkhurst [33], who revealed that out of this state of hypervigilance, participants feel the need to increase their sense of security. Strategies for doing so include both institutional changes (such as increased security guards and surveillance cameras) as well as enhanced personal preparedness (such as changing one’s approach or physical intervention techniques). Our analysis supports previous studies suggesting that heightened vigilance in combination with fear might affect the personal lives of participants [33, 36]. We demonstrated that this might be particularly true for participants who were attacked from behind. An essential distinction emerging from our phenomenological study is whether or not the assault was anticipated. If there was an element of predictability in the assault, hypervigilance is limited to the workplace, whereas if the participant was attacked by surprise, hypervigilance and fear can extend to his or her personal life and translate into a constant perceived vulnerability. The results also suggest that hypervigilance is a strategy developed by participants to prevent additional violence. Thus, the present results add an in-depth perspective to previous quantitative researchfindings.

4.2 Fear

In the literature, few authors have made a distinction between a specific fear of the aggressor and a generalized fear of all patients after a SVA. Fernandes and colleagues [37] established this categorization, indicating that 73% of emergency responders developed a fear of patients as a result of workplace violence. Specifically, 24% of respondents exhibited a specific fear of the aggressor, 35% feared patients with the potential for aggression, and 14% feared patients in general. The present study adds a comprehensive perspective, revealing that more participants report a specific fear of the aggressor rather than of patients in general. In line with the results of our study, we hypothesize that a majority of employees in healthcare settings are caring in their work and thus, following a SVA, their fear is limited to their aggressor. However, some atypical cases were observed in this study: two caring participants did express a generalized fear following the SVA. Overall, it seems that workplace violence interrupts or compromises provided care [38], which corresponds to the consequences described in the two types of fear identified in this study.

Furthermore, previous studies have documented a specific fear of the aggressor. Participants reported making efforts to avoid their aggressor [39, 40], whereas some refused to care for him or her [32] because of fear [15, 41]. Our results also confirm the existing literature to the extent that the healthcare workers’ perceptions of the aggressor are modified in the wake of a SVA. Some healthcare workers feel a sense of powerlessness in the presence of the patient and a lack of empathy for him or her [15, 42]. Our phenomenological study also reveals an important finding that has not been raised in previous research, which is the polarization of attitudes and behaviors following the SVA. In fact, a healthcare worker might withdraw from the treatment of the aggressor, while simultaneously increasing his or her engagement with other patients. This may be considered as a strategy for preventing subsequent violence on the job.

In relation to the presence of generalized fear, withdrawal and divestment among healthcare workers has been demonstrated in previous studies. Some studies corroborate our observation that the experience of violence might lead to more negative attitudes toward tasks and patients [24], passivity [43], dehumanization of patient care and a decline in the healthcare worker’s ability to empathize with patients [44]. Among the responses that have been observed are distancing attitudes [45], loss of interest (an “I don’t care” attitude) and a decrease in sympathy toward the patients [46].

Our findings also suggest a clear link between fear and symptoms related to post-traumatic stress disorder, such as those described in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) [47]. Indeed, those we interviewed had experienced— or witnessed— a traumatic event. Thereafter, they tried to avoid memories, thoughts and feelings related to the associated distress or elements that could awaken these memories, thoughts or emotions (e.g., by avoiding the specific aggressor or patients in general). In addition, cognitive and emotional changes were noted following the event, including persistent and exaggerated negative beliefs and expectations about oneself, others or even the world (e.g., the world is described as “dangerous”), persistent cognitive distortions about the causes or consequences of trauma (e.g., altered perceptions of patients), and a waning interest in important activities or reduced participation in these activities (involvement and commitment to one’s work). Interviews also revealed a marked impairment in the level of activation and reactivation associated with the traumatic event or exacerbated after the trauma (e.g. hypervigilance). Finally, the experience of violence has important implications for participants’ social and occupational life, as well as daily practices (e.g., locking oneself in the car). The apparent similarity between the symptoms listed in the DSM-5 and the hypervigilance and fear described in this study has important clinical implications. We hypothesize that healthcare workers who experience patient aggression in the psychiatric sector are at risk of developing post-traumatic stress disorder and thus agree with such suggestions in previous literature [4].

4.3 Caring

Our findings concerning caring partially support Delaney and Johnson [48], who demonstrated that the commitment of nurses to their patients acts as a protective element that may prevent violence. Indeed, healthcare workers generally seek to create meaningful and stimulating connections with their patients, characterized by activeness, compassion, respect, honesty and authenticity, features that have been conceptualized as “authentic engagement” [49, 50]. Our results are also consistent with Schaftenaar’s andVan der Helm’s [51] concept of “connected care.” These authors highlighted the importance of a specific form of care marked by the stable and reliable presence of a caregiver, which permits the patient to feel that he or she is being taken into consideration, listened to and understood. This makes it possible to establish a relationship of trust— a genuine and mutual connection— that is crucial for the patient’s development and improvement. The concept of caring that emerged in this study resembles the theory of caring that is prevalent in nursing science [52]. Indeed, caring is an interpersonal process involving sensitivity and moral and emotional maturity on the part of the caregiver and has the effect of promoting well-being for both the patient and the caregiver. The present study offers some nuances in this regard by highlighting various aspects of caring (i.e., patient-centered care, dynamic and interactive relationships, and trust), which can be perceived as continuums, and by noting that this outlook can be present among all healthcare workers, and not only nurses. Similarly, Warelow, Edward and Vinek [27], suggested that caring tends to fluctuate between two poles (caring and non-caring) and that a nurse’s work fits somewhere along the continuum. In addition, changes along the caring continuum may be explained by the disenchantment process, as suggested by Tzafrir, Enush and Gur [53]. After rationalizing an experience of aggression, a caregiver may become aware of the gap between their expectations and reality. This realization may generate negative emotions that trigger a cognitive revaluation of their values and norms, which in turn changes their professional conduct. The authors explain that caregivers may go back and forth in the stages of the disenchantment processes at any time during their career, which may be linked to constant movement on the caring continuum.

More tentatively, interesting connections can be made to Akerstrom’s [54] claim that violence in homes for the elderly is not described and named as such. In fact, caregivers find alternative ways to describe their experiences with aggression, which indicates both a level of acceptance as well as the desire to continue their work without questioning their caring role. Employees of a psychiatric hospital and the nature of their work are also defined in terms of their caring role. From this perspective, we might read participants’ specific fear of their aggressor as conducive to preserving their caregiver identity. It would be very difficult to acknowledge that one works with a clientele that has an inherent potential for aggression. To do so would be to question one’s identity as primarily that of a caregiver. In contrast, for employees who do not demonstrate this caring, a SVA exposes their work as being dangerous and causes them to withdraw from patients and to consider changing jobs completely. This hypothesis may shed some light on our findings that caring employees are wary of their aggressor exclusively, whereas those that show little to no caring disinvest from their relationships with all patients following an SVA.

4.4 Limitations of the study

As is the case with all interview-based research, it is possible that participants reported specific situations in an altered form. These modifications are not as limiting as they appear, as the phenomenological approach is interested in the participants’ lived experiences as they remember them and the influence that these perceptions have on behavior. There is nevertheless a possibility that participants generated responses deemed desirable to the researcher or necessitated by specific social standards. With that said, the attitude of the researcher was one of openness and discovery; we thus believe that the participants had little motivation to modify or actively control their descriptions.

Another potential limitation lies in the subjectivity of the researcher. The stage of analysis involving the identification of topics within transcripts may be vulnerable in this regard. One way to minimize this effect is to hold discussions between researchers, as did the two principal investigators of this study, thus utilizing inter-rater reliability [55]. In addition, researchers should ensure that participants are capable of describing their lived experiences and have the necessary communication skills to do so [56].

Finally, comparisons between the current findings and previous research should be made cautiously, as much of the existing literature addresses workplace violence generally rather than in the particular setting of a psychiatric hospital or the mental health sector.

5 Implication for practice

The similarities between previous research, the current results, as well as the symptoms listed in the DSM-5 suggest important clinical and practical implications. These similarities allow for the inference that employees victimized by workplace violence in a psychiatric setting are at risk of developing PTSD, confirming what has previously been suggested in the literature [4 , 58]. They may also be at risk of compassion fatigue, as mentioned by Figley [59]. This concept is defined as an inability to feel empathy towards one’s patients or a lack of interest in doing so.

All of these consequences resulting from type II violence bring about important preoccupations that need to be addressed. The development of prevention and awareness programs as well as reactive programs (intervention and support) is therefore crucial. Based on our results, it may be pertinent for these programs to be based on recognized therapeutic interventions for the treatment of PTSD. Following this model, the programs would target three levels of prevention: 1) primary, in order to reduce the risk of exposure to SVA and prepare workers for potential incidents, 2) secondary, in order to reduce the risk of developing psychological or psychiatric issues following an SVA, and 3) tertiary, in order to reduce the risk that these mental health issues become chronic.

Thus, these programs may limit the extent of the negative impact of SVA on targeted healthcare workers, on the quality of their work with regards to their patients, and on the workplace as a whole. Evidently, it is also crucial for research to continue to address questions related to workplace violence while exploring and specifying the interventions that may improve the outcomes of workers who are violently victimized.

6 Conclusion

The original and innovative aspect of this study is the importance accorded to distinguishing between hypervigilance and fear, and between specific and generalized fear, deepening the understanding of findings from previous studies. To grasp the degree of hypervigilance, it is essential to take into account the concept of anticipation, that is, the extent to which the violence was predictable. This research also reveals the need to consider the concept of a caring continuum in psychiatric settings, and the extent to which it can influence the perception and impact of violence on the victim and his or her work. We can thus ask, what is the origin of caring? Is it learnt on or does it result from a vocation? How might an employer evaluate whether a worker displays caring? This research is essential given that Carlsson [60] claims that people who do not have this outlook (i.e. those who engage in “detached and impersonal care”) can actually contribute to the escalation of anger among patients.

Furthermore, it is necessary to question the impact of violence on caring. Are multiple experiences of violence likely to hinder or inhibit a caring outlook and eventually lead to withdrawal and disengagement? Is what Figley [59] calls “compassion fatigue” also at work in the disengagement of healthcare workers? Just as gratification and positive feedback from patients can ensure job satisfaction and facilitate the ability to persevere at work despite experiencing an SVA, Scott [61] hypothesized that cumulative abuse can lead to emotional exhaustion and depersonalization. This could cause the healthcare worker to withdraw emotionally and physically from the patient, compromising the provision of patient-centered care. As to not perpetuate a blaming approach, it could be that the experience of an SVA results in healthcare workers engaging differently in their work, depending on their ability to continue the same ways of work in which they engaged before.

Finally, the literature shows that the impact of SVAs might be influenced by the sex [62] and the profession [63] of the healthcare worker. Although the methodology of the current study did not allow for a sample large enough to examine such differences, we believe that this aspect should be further explored.

Conflict of interest

None to report.

Funding

This research was funded by La Fondation de l’Institut universitaire en santé mentale de Montréal (IUSMM) through a grant accorded to Stéphane Guay and André Marchand.

References

Foley

, Rauser

. Evaluating progress in reducing workplace violence: Trends in Washington State workers’ compensation claims rates, 1997-2007. Work 2012;42:67–81.

Chen

W-C

, Hwu

H-G

, Lin

Y-P

, Guo

Y-LL

, Su

T-S

, Wang

J-D

. Workplace violence from psychiatric patients. Journal of Occupational Safety and Health 2010;18:163–76.

Lanctôt

, Guay

. The aftermath of workplace violence among healthcare workers: A systematic literature review of the consequences. Aggression and Violent Behavior 2014;19:492–501.

Gates

, Gillespie

, Succop

. Violence against nurses and its impact on stress and productivity. Nursing Economic$ 2011;29(2):59–67.

Barling

, Dupré

, Kelloway

. Predicting worplace aggression and violence. Annual Review of Psychology 2009;60:671–92.

Dupré

, Barling

. Predicting and preventing supervisory workplace aggression. Journal of Occupational Health Psychology 2006;11(1):13.

Miedema

, Hamilton

, Lambert-Lanning

, Tatemichi

, Lemire

, Manca

, et al. Prevalence of abusive encounters in the workplace of family physicians: A minor, major, or severe problem? Canadian Family Physician 2010;56(3):101–8.

de Léséleuc

. La victimisation criminelle en milieu de travail. Ottawa: Statistique Canada, 2007.

Hesketh

, Duncan

, Estabrooks

, Reimer

, Giovanneti

, Hyndman

, et al. Workplace violence in Alberta and British Columbia hospitals. Health Policy 2003;63:311–21.

10.

Merchant

, Lundell

. Workplace violence intervention research workshoAmerican Journal of Preventive Medicine 2001;20(2):135–40.

11.

Tragno

, Duveau

, Tarquinio

. Les violences et agressions physiques au travail: Analyse de la littérature. Revue Européenne de Psychologie Appliquée 2007;57:237–55.

12.

AbuAlRub

, Al-Asmar

. Physical violence in the workplace among Jordanian hospital nurses. Journal of Transcultural Nursing 2011;22(2):157–65.

13.

Magnavita

, Heponiemi

. Workplace violence against nursing students and nurses: An Italian experience. Journal of Nursing Scholarship 2011;43(2):203–10.

14.

Menckel

, Carter

, Viitassara

. Violence towards caregivers of person with developmental disabilities. Work 2000;15:3–8.

15.

Lundström

, Saveman

B-I

, Eisemann

, Aström

. Prevalence of violence and its relation to caregivers’ demographics and emotional reactions - an explorative study of caregivers working in group homes for persons with learning disabilities. Scandinavian Journal of Caring Sciences 2007;21:84–90.

16.

Atan

, Arabaci

, Sirin

, Isler

, Donmez

, Guler

, et al. Violence experienced by nurses at six university hospitals in Turkey. Journal of Psychiatric and Mental Health Nursing 2012;20:882–9.

17.

Talas

, Kocaöz

, Akgüç

. A survey of violence against staff working in the emergency department in Ankara, Turkey. Asian Nursing Research 2011;5(4):197–203.

18.

McGovern

, Kochevar

, Lohman

, Zaidman

, Gerberich

, Nyman

, et al. The cost of work-related physical assaults in Minnesota. Health Services Research 2000;35(3):663–86.

19.

Samir

, Mohamed

, Moustafa

, Saif

. Nurses’ attitudes and reactions to workplace violence in obstetrics and gynaecology departments in Cairo hospitals. Eastern Mediterranean Health Journal 2012;18(3):198–204.

20.

Fry

, O’Riordan

, Turner

, Mills

. Survey of aggressive incidents experienced by community mental health staff. International Journal of Mental Health Nursing 2002;11:112–20.

21.

Demir

, Rodwell

. Psychosocial antecedents and consequences of workplace aggression for hospital nurses. Journal of Nursing Scholarship 2012;44(4):376–84.

22.

Fujishiro

, Gee

, de Castro

. Associations of workplace aggression with work-related well-being among nurses in the Philippines. American Journal of Public Health 2011;101(5):861–7.

23.

Eker

, Özder

, Tokaç

, Topçu

, Tabu

. Aggression and violence towards health care providers, and effects thereof. Archives of Psychiatry and Psychotherapy 2012;4:19–29.

24.

Arnetz

, Arnetz

. Violence towards health care staff and the possible effects on the quality of patient care. Social Science and Medicine 2001;52:417–27.

25.

Fernandes

CMB

, Raboud

, Christenson

, Bouthillette

, Bullock

, Ouellet

, et al. The effect of an education program on violence in the emergency department. Annals of Emergency Medicine 2002;39:47–55.

26.

Gates

, Fitzwater

, Meyer

. Violence against caregivers in nursing homes: Expected, tolerated, and accepted. Journal of Gerontological Nursing 1999;25(4):12–22.

27.

Warelow

, Edward

K-L

, Vinek

. Care: What nurses say and what nurses do. Holistic Nursing Practice 2008;22(3):146–53.

28.

Giorgi

. De la méthode phénoménologique utilisée comme mode de recherche qualitative en sciences humaines: Théorie, pratique et évaluation. In: Poupart

, Deslauriers

, Groulx

, Laperrière

, Mayer

, Pires

, editors. La recherche qualitative: Enjeux épistémologiques et méthodologiques. Gäétan Morin: Boucherville; 1997. pp. 341–64.

29.

Fortin

, Côté

, Fillion

. Fondements et étapes du processus de recherche. Montréal: Chenelière-Éducation, 2006.

30.

Benner

. The tradition and skill of interpretive phenomenology in studying health, illness, and caring practices. In: Benner

, editor. Interpretive phenomenology: Embodiment, caring and ethics in health and illness. Thousand Oaks: Sage Publications; 1994.

31.

Karlsson

. Psychological qualitative research from a phenomenological perspective. Stockholm, Sweden: Almqvist & Wiksell, 1993.

32.

McKenna

, Poole

, Smith

, Coverdale

, Gale

. A survey of threats and violent behaviour by patients against registered nurses in their first year of practice. International Journal of Mental Health Nursing 2003;12(1):56–63.

33.

Kindy

, Petersen

, Parkhurst

. Perilous work: Nurse’s experiences in psychiatric units with high risks of assault. Archives of Psychiatric Nursing 2005;19(4):169–75.

34.

Crilly

, Chaboyer

, Creedy

. Violence towards emergency department nurses by patients. Accident and Emergency Nursing 2004;12:67–73.

35.

Cheung

, Bessell

, Ellis

. The final straw. Australian Nursing Journal 2004;11(11):17–9.

36.

Levin

, Hewitt

, Misner

. Insights of nurses about assault in hospital-based emergency departments. Image Journal of Nursing Scholarship 1998;30(3):249–54.

37.

Fernandes

CMB

, Bouthillette

, Raboud

, Bullock

, Moore

, Christenson

, et al. Violence in the emergency department: A survey of health care workers. Canadian Medical Association Journal 1999;161(10):1245–8.

38.

Canton

, Sherman

, Magda

, Westra

, Pearson

, Raveis

, et al. Violence, job satisfaction and employment intentions among home healthcare registered nurses. Home Healthcare Nurse 2009;27(6):364–73.

39.

Henderson

. Nurses’ experience of workplace violence: Towards effective intervention. British Columbia: Focus on Tomorrow, 2010.

40.

Needham

, Abderhalden

, Halfens

RJG

, Dassen

, Haug

H-J

, Fischer

. The impact of patient aggression on carers scale: Instrument derivation and psychometric testing. Scandinavian Journal of Caring Sciences 2005;19:296–300.

41.

C-H

, Wang

T-F

, Liu

C-Y

. Psychiatric nurses’reactions to assault upon them by inpatients: A survey in Taiwan. Psychological Reports 2007;100:777–82.

42.

Aström

, Karlsson

, Sandvide

, Bucht

, Eisemann

, Norberg

, et al. Staff’s experience of and the management of violence incidents in elderly care. Scandinavian Journal of Caring Sciences 2004;18:410–6.

43.

Levin

, Hewitt

, Misner

, Reynolds

. Assault of long-term care personnel. Journal of Gerontological Nursing 2003;29(3):28–35.

44.

Zimbardo

, Maslach

. Depersonalisation. In: W

B.B.

, editor. International encyclopedia of psychiatry, psychology, psychoanalysis and neurology. 4. New York: Human Sciences Press; 1978. pp. 1–13.

45.

Gates

. Burgers or bruises? American Journal of Nursing 2004;104(9):13.

46.

Bimenyimana

, Poggenpoel

, Myburg

, Van

Niekerk V

. The lived experience by psychiatric nurses of aggression and violence from patients in a Gauteng psychiatric institution. Curationis 2009;32(3):4–13.

47.

American Psychiatric Association. Diagnostic and statistical manual of mental disorders 5th ed. Washington, DC, 2013.

48.

Delaney

, Johnson

. Metasynthesis of research on the role of psychiatric inpatient nurses: What is important to staff? Journal of the American Psychiatric Nurses Association 2014;20(2):125–37.

49.

Finfgeld-Connett

. Model of Therapeutic and Non-Therapeutic Responses to Patient Aggression. Mental Health Nursing 2009;30:530–7.

50.

Van Bogaert

, Wouters

, Willems

, Mondelaers

, Clarke

. Work engagement supports nurse workforce stability and quality of care: Nursing team-level analysis in psychiatric hospitals. Journal of Psychiatric and Mental Health Nursing 2012.

51.

Schaftenaar

, Van der Helm

, editors. The enduring myth of the violent client’, living group climate and aggression in a forensic psychiatric setting. 8th European Congress on Violence in Clinical Psychiatry; 2013; Ghent, Belgium: KAVANAH.

52.

Finfgeld-Connett

. Meta-synthesis of caring in nursing. Journal of Clinical Nursing 2008;17:196–204.

53.

Tzafrir

, Enosh

, Gur

. Client aggression and the disenchantment process among Israeli social workers: Realizing the gaQualitative Social Work 2015;14(1):65–85.

54.

Akerström

. Slaps, punches, pinches - but not violence: Boundary-work in nursing homes for the elderly. Symbolic Interaction 2002;25(4):515–36.

55.

Giorgi

, Giorgi

. Phenomenology. In: Smith

, editor. Qualitative psychology: A practical guide to research methods. London: Sage; 2003.

56.

Hycner

. Some guidelines for the phenomenological analysis of interview data. Human Studies 1985;8:279–303.

57.

Wykes

, Whittington

. Prevalence and predictors of early traumatic stress reactions in assaulted psychiatric nurses. Journal of Forensic Psychiatry 1998;9:643–58.

58.

Whittington

, Wykes

. Staff stain and social support in a psychiatric hospital following assault by a patient. Journal of Advanced Nursing 1992;17:480–6.

59.

Figley

. Compassion fatigue as secondary traumatic stress disorder: An overview. In: Figley

, editor. Compassion fatigue: Coping with secondary traumatic stress disorder in those who treat the traumatized. New York: Brunner-Routledge; 1995. pp. 1–20.

60.

Carlsson

, Dahlberg

, Ekebergh

, Dahlberg

. Patients longing for authentic personal care: A phenomenological study of violent encounters in psychiatric settings. Issues in Mental Health Nursing 2006;27:287–305.

61.

Scott

, Ryan

, James

, Mitchell

. Perceptions and implications of violence from care home residents with dementia: A review and commentary. International Journal of Older People Nursing 2010;6:110–22.

62.

Guay

, Goncalves

, Jarvis

. Verbal violence in the workplace according to victim’s sex - a systematic review of the literature. Aggression and Violent Behavior 2014;19:572–8.

63.

Virkki

. Habitual trust in encountering violence at work: Attitudes towards client-violence among Finnish social-workers and nurses. Journal of Social-Work 2008;8(3):247–67.