Threats and Violence in the Care Sector

Abstract

Workplace violence is a complex occupational hazard that health care staffs are facing in today’s work environment. This study examines the safety measures recommended by staff in occupational injury reports filed in the wake of violence- and threat-related injuries. The results suggest that measures relating to psychosocial factors and structural factors such as adequate staffing are much more important than surveillance, staff training, and penal sanctions.

Keywords

mental health and crime crime in complex organizations violent behavior critical criminology crime/delinquency theory crime prevention law enforcement/security

If an employee in Sweden is injured, for example, as a result of threats or violence, the employer is required to submit an occupational injury report. This report may include proposed measures to prevent a repeat of the incident. This article analyses the measures proposed by employees in the care sector in the years 1987, 1997, and 2007.

In 2010, approximately 2,500 occupational injury reports relating to threats and violence were submitted in Sweden. One third of these were submitted by the care sector—whose employees represent one of the most victimized occupational groups (Arbetsmiljöverket, 2011). Since the passing of the first health and safety at work legislation in 1905, discussions of how to produce a good work environment have been informed by the view that it is the employees who have the knowledge required to produce such a work environment (Hydén, 2004, p. 134). Official statistics relating to accidents at work, which include threats and violence, were first published in 1906 and are collated in the information system on occupational injuries (The Swedish Information System on Occupational Accidents and Work-related diseases [ISA]), which is among other things intended to provide a basis for efforts to prevent injuries at work (Arbetsmiljöverket, 2011, p. 28).

The focus of the current article is directed at the measures proposed by those exposed to violence. One of the advantages associated with studying occupational injury reports is that they provide an opportunity to examine the views regarding preventive measures held by the victims themselves. In addition, they have been collected for a long period, which allows us to examine whether there have been changes over time. The current study focuses on three specific years: 1987, 1997, and 2007.

The Health Care Industry in Sweden

Everyone in Sweden has equal access to health care services under a largely decentralized, taxpayer-funded system. The Swedish health care system is mainly government funded although private health care also exists. The state finances the bulk of health care costs, with the patient paying a small nominal fee for examination. The state pays for approximately 97% of medical costs. When visiting a hospital, the entrance fee covers all specialist visits the doctor deems necessary, like X-ray, specialists, surgery operations, and so on. The same fee is levied for ambulance services.

Like many other countries, Sweden faces numerous challenges, such as funding, quality, and efficiency of its health care services. During the 1990s, there were more or less continuous reductions in staffing levels in the health and geriatric care sectors. The Swedish population of 2001 was being cared for by 15% fewer care staff than the population of 1990, despite being the same size and getting older. Larsson (2004) argues that during the 1990s, the work environment of those in municipal employment underwent a substantial deterioration. The work environment of carers in the geriatric sector has deteriorated by comparison with the beginning of the 1990s (Gustafsson & Szebehely, 2005). Also since the beginning of the 1990s, Statistics Sweden’s level-of-living surveys (The Swedish Living Conditions Surveys) show an increase in levels of work-related violence within the female-dominated sectors of the school system and care provision. The increase is explained as being due to an increased reporting propensity and changes in the work environment, and it has occurred at the same time as corresponding increases are not found in other vulnerable sectors such as the retail sector, the police, and security services (Estrada, Nilsson, Jerre, & Wikman, 2010).

Current Study Focus

As levels of violence against care-sector employees have increased, together with the amount of attention focused on this violence, views on how this violence should be dealt with have also undergone certain changes. A study of proposed measures to combat violence at work described in trade journals shows that it became more common over time for care-sector employees to be portrayed as crime victims who require redress in the form of action on the part of the justice system rather than preventive and rehabilitative measures on the part of the employer (Wikman, 2012). The study also showed a trend toward juridification and responsibilization. The question, then, is whether these changes are also reflected in the measures proposed by the victims of violence themselves. In order to be able to problematize the question of whether situational conditions and proposed measures have changed over time, it is necessary to consider whether the violence in question is occurring in new situations or whether the character of the violence has changed, for example, by becoming more serious. The article’s objective is to analyze and categorize the measures proposed by victims of violence themselves and to discuss changes over time. The following research questions are addressed:

Research Question 1: How has the violence reported changed over time?

Research Question 2: Does the violence occur in new situations, and has the nature of the violence changed?

Research Question 3: Has the view of the perpetrator changed?

Research Question 4: What measures do the victims of violence propose in their occupational injury reports? Have the proposed measures changed over time?

The article proceeds with a review of the research on measures to combat violence at work, followed by a presentation of the study’s theoretical points of departure and a short description of the interpretational framework employed in the analysis. The results are presented in two stages: first in the form of quotes drawn from the occupational injury reports and second in the form a quantitative overview of the frequency with which different types of measures is proposed, and trends over time. Finally, the main points are summarized in a concluding discussion.

Research on Measures to Combat Violence at Work

Although prevalence levels as well as the character and intensity of experience vary substantially between organizations and between countries, the presence of violence in some form appears to be omnipresent and endemic within the health service sector (International Labor Office/World Health Organization/International Council of Nurses/Public Services International, 2000). A substantial amount of the research on violence at work has been focused on issues relating to the extent of this violence and the risk situations in which it occurs (Braverman, 2002; Chappell & Di Martino, 2006).

A review (Hills & Joyce, 2013) found a scant literature on the prevention and management of workplace aggression in clinical medical practice, with only a small body of research studies relating to nursing or mixed professional groups or settings, such as in psychiatry or emergency departments. While the review has identified broad advocacy for the integration of diverse approaches to the preventions and management of workplace aggression in health settings, most has been based on expert and theory. Hills and Joyce (2013) notes that systems for reporting workplace aggression may be valuable in providing detailed records of individual incidents of violence, patterns of violence across the organization, and a basis for implementing targeted prevention.

One literature study (Wikman, Estrada, & Nilsson, 2010) shows that some types of situation are known to be able to produce increased levels of worry and aggression. These are situations involving the communication of negative information or decisions, the denial of access to service, queues, staff experiencing stress, and “general disorder.” A large proportion of the violence that occurs within the care sector does so in connection with activities such as washing, dressing or undressing, feeding, and the like. Physical conditions such as poor air quality, lighting, and uncomfortable temperatures can also produce frustration, which may in turn lead to aggression and violence (Wikman et al., 2010).

There is also an antagonism between the need to be close to the patients, providing care quickly, and the need for caution among staff members. In the emergency room, staffs are required to help those in need of assistance as quickly as possible at the same time as dangerous and sharp objects, such as syringes, have to be identified. A high workload can also increase the level of risk, by making it impossible to act with the required caution. Cuts in the form of redundancies and/or reduced staffing levels per client/patient have also been shown to produce risks. Having little control, in the sense of having few opportunities to influence one’s work situation, low levels of knowledge, and little experience are also associated with an increased risk of exposure to violence in the workplace (Wikman et al., 2010).

A number of researchers have noted that knowledge on measures to combat violence at work is more limited (Bowie, 2010; Mayhew, 2004; Paterson, Leadbetter, Miller, & Bowie, 2009). One of the problems identified in a review of the existing knowledge is that few measures to combat violence at work have been evaluated and shown to be effective (Wikman et al., 2010). This is true both for direct responses to violence and for long-term crime prevention strategies (Braverman, 2002; Chappell & Di Martino, 2006).

The importance of a preventive focus that avoids reactive control is often emphasized, and references to the complexity of the causal factors that can lead to violence, and how these should be dealt with, are also recurrent themes. All things considered, the literature shows that factors related to the way work is organized are of major significance for understanding why violence occurs (Bowie, 2010; Jones, Robinson, Fevre, & Lewis, 2011).

An evaluation of support provision for victims of workplace crime shows that support from colleagues and making time to talk about what has happened at the workplace are both measures that victimized staff members perceive as being helpful (Carlsson & Wennerström, 2010; Paterson, Leadbetter, & Bowie, 1999).

A Swedish representative interview study of 400 health care employees from the early 1990s found that 45% of the employees stated that their employers should do more to deal with “aggressive” patients (Åkerström, 1993). The measures that were most often requested took the form of psychological or medical training in order to understand the patients’ aggressiveness. At the same time, there were substantial differences in the nature of the measures desired across different workplaces. Emergency room staff stated that both technical solutions and training measures were needed, whereas staff at outpatient clinics, mental hospitals, and nursing homes to varying extent noted the need for training in group-based methods and self-defense (Åkerström, 1993, p. 46).

What Is the Function of an Occupational Injury Report?

The occupational injury report form in Sweden includes a special box where the injured party is encouraged to suggest measures that could prevent the injury occurring again. Petersson (2012), who has studied occupational injury reports relating to chronic exhaustion, argues that although there is legislation regulating what is to be deemed an occupational injury, occupational injuries are nonetheless socially constructed. According to Petersson, the reports appear to constitute a way of “saying stop” and may be regarded as testimony to the existence of detrimental conditions at the workplace. She presents examples of people who have written that they are submitting the occupational injury report in order to make others aware of the situation at their workplace and because doing so may lead to changes not only at their own workplace but across an entire sector (Petersson, 2012).

Åkerström (1997) has shown that filing an occupational injury report may be viewed as an act of duty in certain care-sector cultures. Filing such a report can constitute a means of drawing attention to and documenting the extent of the workload. Petersson (2012) notes that the arguments made by those filing such reports show that they expect the state to take responsibility in a way extends beyond the economic and insurance-related relationship, in which the state is expected to assume responsibility for the work environment. Those filing the occupational injury reports express a vision of occupational injury insurance that conveys a belief in the possibilities open to the state and political decision makers.

Theoretical Points of Departure

There are a number of, often tacit, conceptions that constitute the prerequisites for a given countermeasure to appear rational. In the current context, these conceptions include views on the causes of crime and on the factors that it is possible to affect (Giertsen, 1994; Sahlin, 2000). Different academic disciplines, for example, are associated with different traditions regarding the countermeasures that are viewed as possible or reasonable, depending on the study object in focus. The focus of psychologists is primarily directed at individual behavior, whereas sociological research focuses on the actions of groups, which also affects the type of countermeasures that are “available” (Paterson et al., 2009). The prevention of problems is often equated with obstructing their onset (Sahlin, 2000, p. 24). Sahlin argues, however, that there is both a mutual dependence and a relative independence between problems and countermeasures, which means that preventive efforts serve functions other than those described in their publicly stated motivations at the same time as the measures introduced can influence the way a problem and its causes are perceived (Sahlin, 2000, p. 80).

This double function is illustrated in a survey of occupational injury reports conducted by Åkerström (1997), in which she examines local practices in the form of approaches to the tolerance, handling, and reporting of violence at different workplaces. Åkerström shows that is was more common to file occupational injury reports in the mental health care sector than in geriatric care, emergency care, or outpatient care. One explanation posited by Åkerström was that the injuries suffered by mental health care workers were more serious. Another was that the staff wanted to demonstrate that their workload was high. According to Åkerström (1997), occupational injury reports may serve not only a manifest function, that is, that of applying for financial compensation, but also a latent, rhetorical function, for example, as a means of documenting the extent of the workload.

Another of Åkerström’s (2002) findings was that the problems associated with categorizing which incidents should and should not be conceptualized as constituting violence in the care sector have more to do with the categorization of people than of events. In the current study, it is assumed that the factors of significance for the measures proposed are where and in what situation the violence occurred, and the character of the violence. In addition, there is an interaction between these factors and the view of the perpetrator as being either a rational “bad guy,” and thus an individual who may be receptive to punishment as a form of deterrent, or a “poor thing,” and thus immune to the influence of punishment and instead in need of care or treatment. If a crime can be regarded as a symptom of some form of illness, the incentive to punish disappears, since the offence’s moral dimension, as the intentional infliction of an injury on another, is de-emphasized. This can be contrasted with the situation where the perpetrator is instead viewed as a rational actor who alone bears responsibility for the act, and who can thus take the consequences of his or her actions in the form of imprisonment or some other form of punishment. The concepts of the “poor thing” and the “bad guy” are employed in this study to correspond to the classical theme of treatment contra punishment, which are based on two different and mutually incompatible systems in the form of medicine and the law. The dichotomy is used to capture the difference between two stereotypes that may be viewed both as an expression of and as an argument for a given policy. As such, the dichotomy is not absolute but rather constitutes one element in the construction of social problems.

Framing

The framing concept was first employed by Goffman (1974) and is often used as a means of describing the “policy making” of social movements (Paterson et al., 2009). “Framing” is a dynamic process that forms the collective frames of action that people create in order to express their opinions (Benford & Snow, 2000). In the context of my studies of existing research, for example, I have made active choices and interpreted that which is presented and “framed” as being more relevant than other content. Tentative interpretations are then compared with specific examples. When the same frames recur, these acquire a self-evident status as a form of “genre.” The frames represent approaches and methods that implicitly and explicitly constitute perspectives on how the problem of violence is defined, and they presuppose certain views on the causes of violence. The theory emphasizes a dynamic process that I would argue is well suited to studying the construction of views regarding the offender and the countermeasures that follow as a consequence of these views. This is reminiscent of what others have labeled boundary work (see Åkerström, 2002; Gieryn, 1983). Unlike boundary work, however, which shows which actions or events are included within the violence concept, the frames employed in the current study focus on the way the people who commit the acts of violence are viewed. The study examines who the boundary relates to rather than where the boundary is drawn. I will return to the methodological issue of how the interpretational frames have been applied to the study material in subsequent sections.

Data and Method

Data

In order to be able to describe changes over time, reports were selected from the years 1987, 1997, and 2007. The sampling frame comprised all occupational injury reports from members of the groups working in the care sector that occupational injury statistics show to be most exposed to threats and violence (ambulance paramedics and careers, at-home support staff, child and youth carers, habilitation assistants, home carers, personal assistants, mental health nurses, nursing assistants, and assistant nurses). All the reports from 1987 were included in the sample, together with every third report from 1997 and every second report from 2007. The choice of these proportions from the respective years was made in order to produce a sample for each year that was large enough to be broken down into subcategories.

The total number of occupational injury reports varies substantially between the different years covered by the study (see Table 1), for which there are several explanations. The contents and methods used in connection with the published statistics have been largely the same since 1979, but the statistical systems employed during the periods 1980–1993, 1994–2002, and 2003–2008 are not entirely comparable with one another. The categorization of branches and occupations has changed over time, for example. In 1992, a 14-day period of sick leave was introduced, during which the employer pays the employee’s wages. This change has probably meant that occupational injuries leading to a short period of sick leave are not reported to the same extent as previously. In 1993, a day with no sick pay was introduced at the beginning of a period of sick leave which may have further reduced the propensity to file reports in connection with short periods of sick leave. The number of less serious incidents is assumed to be higher in 1997 than in the other years, since in that year, accidents at work that did not involve a subsequent absence from work (such as dental injuries, hearing loss, and psychological reactions to, e.g., threats, robberies, etc.) were also registered (Wikman, 2012).

Table 1.

Number of Occupational Injury Reports at Different Stages of the Sampling Process (ISA 1987, 1997, and 2007).

Care Sector Staff	1987	1997	2007	Total
Total number of reports	587	1,579	779	2,945
Sample	587	526	389	1,502
Following removal of nonrelevant cases	531	506	368	1,405
Nonminor or serious violence	423	273	231	927
Reports of nonminor or serious violence, which include proposed countermeasures	128	132	144	404

Note. ISA = The Swedish Information System on Occupational Accidents and Work-related diseases.

Some research has found differential predictor variables depending on whether the rate or the severity of violence. The rate of violence is much more consistently reported, with less consideration placed upon severity levels (Welsh, Bader, & Evans, 2013). One of the most common ways of defining violence is to only consider forms of criminal violence and to argue that violence is the use of force that has been prohibited by law. However, while “violence” might conventionally connote physical attack, the notion of physical violence, as well as physical injuries, represents a surprisingly broad spectrum of incidents (Schindeler, 2013, p. 8; Waddington, Badger, & Bull, 2005, p. 149). However, in order to avoid problems resulting from the tendency for certain years to include more reports involving “minor violence” than others, since it is reasonable to assume that reports relating to this type of violence are those that are most affected by changes to reporting routines, and so on, the analyses that focus on changes over time will be based on reports relating to nonminor and serious incidents of violence.

The reports in this study relate to acts that may give rise to criminal sanctions.¹ All the reports that do not do so have been excluded from the sample. In a small number of cases,² for example, when the injured party has not returned to work, the safety officer or another workplace representative may have made the report, but otherwise it is the injured employee himself or herself who has signed the report and may therefore be assumed to have completed it.

The advantage of using this material is that it provides knowledge on the countermeasures that have been proposed over time by care staff who have themselves been exposed to threats and violence, together with a deeper understanding of the context in which the violence occurred. At the same time, there are disadvantages in that the material does not provide an opportunity to ask follow-up questions, as would be the case with interview data, and the information provided is at times quite sketchy.

Operationalizations

Countermeasures

More than one measure³ can be proposed on the occupational injury report form. Each countermeasure has been coded individually. The label psychosocial work environment is used to refer to measures involving changes in staffing levels, queuing, working alone, opening times, and working at night. More resources and guidance and crisis management relate to measures that fall under the heading psychosocial work environment, but proposals relating to these particular areas have also been coded separately. The measure referred to as medication of patient involves a focus on adapting the patient to the environment by increasing medication or changing the diagnosis, care plan or the form in which care is provided. Informational work refers to training measures at the group level, cataloguing problems, appointing working groups, or updating routines.

The measures labeled physical work environment focused on work routines/tasks involve adapting the premises to the nature of the work, visiting rooms for taking care of patients, technical solutions, and segregating or locking patients in. Group techniques have been coded as self-defense training.

Physical work environment with a focus on security may refer to bulletproof glass, alarms, reserve exits, routines for handling cash, and surveillance cameras monitored by staff. In the active surveillance category, the surveillance cameras are instead monitored by security guards or the police. The police/police report category involves proposals for a greater police presence in the community and calls for more incidents to be reported to the police.

Risk Situations

Only the situation that most closely corresponds to the specific violent incident or that is most clearly described has been coded. The eight risk situations identified relate to mental ill-health, negative decision/reprimand, intimate situations, work-related stress, food situation, dementia, control of unauthorized persons, and cash.

The Nature of the Violence

The violence has been divided into three categories based on its seriousness: minor , nonminor, and serious violence. The minor violence category relates to violence that did not produce visible marks, being pushed, single punches, or kicks that were not to the face. The nonminor violence category includes incidents that produced visible marks, several punches, or kicks to the face. The serious violence category relates to violence that the occupational injury report shows required the victim to seek medical attention (cf. Brå, 2009, p. 6). The coding is based on what is explicitly mentioned in the occupational injury report.

Different Areas of the Care Sector

It is sometimes difficult to know which area of the care sector a given occupational injury report relates to. In cases where this was uncertain, I have consulted with two individuals with experience from emergency, primary, outpatient, and social services care. I use six main categories: Children, Psychiatry, Addiction, Health care centers, Geriatric, and Daily Care for People with Disabilities. In order to examine the level of intercoder reliability, a subsample of 80 occupational injury reports was coded by two individuals. The level of correspondence was no lower than 85% for any of the variables coded.

Method

To provide an overview of the prevalence of the different proposed measures, the study employs quantitative content analysis, a method whose strength is that it provides a good basis for comparisons (Bryman, 2012). The method involves counting the measures described in the occupational injury reports using a coding scheme formulated on the basis of a literature review,⁴ which included 182 Swedish scientific publications on violence at work from the period 1975–2010 (Wikman et al., 2010).

The study’s frames of interpretation have the objective of illustrating themes, concepts, and perceptions that appear repeatedly in the study material. These can then be related to one another, to existing research and to theoretical perspectives and can be organized into and understood as constituting collective frames of action. I thus test whether it is possible to code the measures described in the injury reports in accordance with the frames proposed. The frames have been constructed in connection with the coding of the material and the research review. By moving back and forth in this way between proposed interpretations and the testing of these interpretations, the study becomes both inductive and deductive and the frames constitute both an analytical tool and a result of the analysis.⁵

The Framing of Prevention Work

Table 2 summarizes the results relating to the measures described in the occupational injury reports. The measures have been broken down into three distinct frames: the normalization frame, the control frame, and the deviance frame. The frames at the two extremes—normalization and deviance—differ in their view of the perpetrator of the violence. In the column describing the normalization frame at the left, we can see that it is viewed as “normal” for care-sector clients to occasionally become violent. Viewed in terms of the frame presented on the right-hand side of the table, the violence is viewed as representing such a deviation that the individual must be dealt with outside the institutions of the care sector. The first frame, the normalization frame, has the objective of creating the conditions required to deal with the clients’ and patients’ behavior, that is, the measures are indirect and focused on structures. The deviance frame involves a combination of involving the police and prosecution service and the use of legislation and the justice system as instruments of “crime prevention.”

Table 2.

Analytic Schedule for Frames, Themes, and Countermeasures.^a

Normalization Frame (Perpetrator Viewed as a “Poor Thing”)		Control Frame (Perpetrator Viewed as Potentially Both a “Poor Thing” and a “Bad Guy”)		Deviance Frame (Perpetrator Viewed as a “Bad Guy”)
Theme	Countermeasure	Theme	Countermeasure	Theme	Countermeasure
Psychosocial work environment	More resources (time, staff, and training)	Exclusion	Diagnosis, medicating, locking patients in	Individualization of responsibility	Police report, fines
Organizational structure	Routines, talking to supervisors or colleagues	Physical routines and configuration of premises	Visiting rooms, technical aids, physical restraints	Juridification	Criminalization
Organizational structure	Routines, talking to supervisors or colleagues	Physical work environment focused on security	Bulletproof glass, locks, reception area, alarms, emergency exits	Self-defense^b	Holding techniques, etc.

^aThe literature includes different proposed classifications of existing crime prevention measures, for example, Balvig (1979–1980), Christie (1999), and Sahlin (2000). I have drawn my inspiration from all of these, but Table 2 differs slightly since the present study proceeds from a victimological perspective, in which it is the measures proposed by those exposed to violence that are counted and interpreted. ^bSelf-defense training is the measure that has been most difficult to assign since it may be used both proactively and reactively. I have come to the view that it fits better among the measures that are linked to the view of the individual as responsible. Further, the need for defense implies that a violent situation has already escalated, which sets the measure a little apart from the measures included in the more preventively focused normalization frame.

The control frame represents an intermediate position that functions as a control point, at which the perpetrator may be excluded from the relevant care-sector territory or area of responsibility either by denying the individual access at an early stage by means of surveillance or at a later stage either by defining the patient as suffering from an illness that falls outside the operational definitions of a specific form of care provision or by “disarming/calming” the patient by means of medication.

In the following section, the frames are illustrated in the form of quotations drawn from the occupational injury reports. The quotations are presented in a form as close to the original as possible, taking into consideration the need to preserve the anonymity of the individuals involved. The year in which the report was made is not presented for the same reason. In those cases where words or phrases have been omitted, this is shown using […]. The number in parentheses is the serial number of the occupational injury report. The presentation begins with quotes exemplifying the normalization frame, which are followed by quotes exemplifying the control frame, and finally the deviance frame. The excerpts from the occupational injury reports presented subsequently include proposed measures, descriptions of measures that have already been taken, and criticisms of failures to implement earlier proposals. The assignment of the quotations to the various frames should therefore be seen as examples, since the quotes often include several measures and as a result could be located in more than one frame. The use of this approach is motivated by the fact that it facilitates the exemplification⁶ of the measures proposed and the contexts in which they occur.

The Normalization Frame

Psychosocial work environment

Within the normalization frame, measures intended to change the psychosocial work environment are central. Here the dominant view is that violence at work should not be dealt with by censuring the individual, the perpetrator, or the staff (Braverman, 2002; Paterson et al., 2009). Violence should instead be viewed as a symptom of failures in the way the work is organized at the structural level. The psychosocial countermeasures that are proposed often take the form of guidance and crisis management, sharing knowledge and more resources.

Carer, group home for person suffering from a mental handicap (938)

I was woken by the night officer. She had a problem with a resident who had got hold of a carving knife. Being alone and faced with a knife and feeling responsible for the other seven individuals who live here and then sitting on the phone and getting passed from one person to another is not a humane work situation. Not getting help immediately from at most a single phone call was extremely frustrating. […] When I went home it felt like I didn’t know whether to laugh or cry. I felt enormously let-down that nobody had informed me about his aggressiveness and that he might turn to a weapon, thought a lot about whether the pay was worth the risks.

Measure to prevent a repeat of the injury: Locked up knives and scissors. Conversation with senior physician. Everything has to work in emergency situations, e.g. calls for assistance to the police. Inform the people who report a new resident how important it is that we are told how violent residents are so that information is not withheld.

It is not uncommon for measures of this kind to be proposed in connection with serious incidents of violence, such as threats involving knives as in the previously mentioned example. The individual who reported the violence expresses anger at not having been given sufficient information about the patient’s condition, but there also appears to be a need to give expression to a sense of inadequacy and frustration over work conditions. Staff members exposed to threats and violence appear to view support from colleagues and setting aside time to talk about what has happened as important. In the next example, we see the need for more resources.

Psychiatric care nurse, hospital department (854)

We were standing in the toilet when the patient twisted a metal object off the wall and threatened to poke my eyes out. I had to twist the object out of the patient’s hand and injured my hand. Got a cut in my little finger.

Measure to prevent a repeat of the injury: Listen to staff who need additional support. Senior physician refused to requisition additional staff. This meant that there was nobody to relieve us. The patient’s anxiety and aggression would have required 2 people. Because of the situation, there was no opportunity for me to leave the ward after I had been injured.

In the previous example, it is clear that the injury could have been avoided if there had been additional staff. The staff member could at least have received medical attention more quickly. Other occupational injury reports describe the burden of the workload as being an area that could be improved, even though in the following example there were three staff members to a single patient.

Psychiatric care nurse, hospital department (1,000)

A patient in restraints was taken up to go to the toilet by three nurses, and is then placed in restraints again. In connection with this, the patient kicked out explosively with both legs, catching me in the abdomen and pushing me backwards very powerfully.

Measure to prevent a repeat of the injury: Reduce the number of beds with the same number of staff.

Organizational Structure

Not uncommonly there are a large number of sectors that have to work together: psychiatric care, habilitation, primary care, mobility services, curators personae and bonis, and needs assessment officers. The list could be made much longer. It is clear from the occupational injury reports that the violence is often due to stress caused by anxiety due to, for example, mobility services that have not arrived, meeting new staff who have not been given sufficient information, and so on. The subsequent quotation also shows that it can be difficult for care staff to know where to turn for help, either for themselves or for the patient. The example is taken from a report made by a temp who was dissatisfied with the environment in which the staff were working:

Medical orderly, care home (1,108)

I asked the patient to stop banging on the door. Then he kicked me several times on the legs and in the crotch. Later on the patient came into the staff room and slapped my face. I was bruised and got a split lip and a nosebleed.

Measure to prevent a repeat of the injury: The work place is suffering from having too few staff, insufficient knowledge and organisational problems. Male staffs are needed. The only “weapon” available to the staff was to assign blame to the patient. No work methods. Poor monitoring and supervision. No possibility for self-defense against violence. The other two women removed the patient from the staff room and threatened, scolded and blamed him, why? Employ trained staff, train the staff in self-defense, demand changes to working conditions. Get yourselves trained! Look at [name] who doesn’t give you any information, Who is the safety officer?

In summary, on the basis of this frame, the violence may be understood as an interaction between an individual and a system that either provokes the violence or allows it to occur.

The Control Frame

Exclusion

Removing violent patients from one’s own workplace can be achieved by defining them as being too ill (violent) for the form of care provided there. Another alternative is to view the patients as being responsible for their actions, despite their illness, in line with the so-called principle of criminal culpability, that is, on the basis of their capacity for being responsible for their actions and then ensuring that they are dealt with by the justice system. In the context of the principle of criminal culpability, Swedish law occupies a rather unique position, since individuals with serious mental illness can be sentenced to criminal sanctions (with the exception of a prison term; Penal Code, chapter 30, section 6; Jareborg & Zila, 2007). Medication in the form of sedatives and being locked into one’s own room may also be viewed as a form of exclusion. Both pathologization, that is, explaining the patient’s violence as being a result of his or her illness in contrast to viewing it as a reaction to detrimental conditions, and the principle of criminal culpability, that is, making the violent patient legally responsible, can be used as a means of moving the patient out of a given care environment.

Home-help assistant (809)

I stood at the sink to put the food on the plates. Then [name] came in clothed in nothing more than a condom. He pushed me up against the wall and shrieked “we’re going to fuck” several times and made powerful sexual thrusting movements against my body. I was gripped by panic. I couldn’t see an escape route and reacted by giving him a slap on the face. [Name] then let go and I ran out of there. [Name] shrieked after me “it’s safe sex”. I needed help!

Measure to prevent a repeat of the injury: Phoned my supervisor but she wasn’t home, phoned my boss but she wasn’t home either, phoned the regular home-help for [name]. Finally! Someone to talk to. I was in complete despair, mostly about having slapped him, but I also needed help to sort out the situation.

My workmate advised me to contact the district nurse at the local health centre. The district nurse didn’t know what I should do but said she would ask a colleague and get back to me by phone. After a while the district nurse called me back and informed me that I should report the incident to the police. I phoned the police’s control centre. Was told that it wasn’t a matter for the police because the incident was not felt to be sufficiently “serious.” The police say that [name] had probably flipped out. I was advised to contact a doctor and cycle down to the station to file a police report. I phoned the district nurse again and asked for help calling a doctor, and was told to ring the national health helpline…

In the previous injury report, there is a lack of clarity as to who is responsible for the client. Is it the home-help service itself, primary care, emergency care, or perhaps even the police? The quotation subsequently exemplifies a desire to disclaim responsibility by moving the client to another part of the care sector:

Personal assistant, at-home care (5)

I am standing in the kitchen looking out of the window. She comes and stands beside me. When I lean forward slightly to look out she become angry, shouts “no” and hits med in the stomach with her hand.

Measure to prevent a repeat of the injury: 19th May: The care recipient is moved to a short stay care home because the accommodation staff are concerned about how to deal with the care recipient’s aggression.

21st May: The care recipient back to her home.

22nd May: Planning meeting with family, doctor, nurses and case officer. Doctor determines that more medication produces side effects in the form of fatigue. Emotional support and sharing knowledge is recommended for the staff, amongst other things on personal interactions. The counseling officer provides tutelage on personal interactions, approaches and neurological injuries.

In this quotation, the care recipient is first judged to be too ill (violent) to be at home, but is then declared healthy 2 days later and sent home again. A meeting is then arranged at which proposals for increased medication are rejected in favor of proposals focused on the need to change the way the staff interact with the care recipient, that is, the incident ends up in the first frame, the normalization frame. In the next example, a man with an intellectual disability is taken into custody by the police and is thereby shifted from the normalization frame to the deviance frame:

Carer, group home for intellectually disabled (561)

He comes and stands in the doorway and shouts “fucking whores and cunts” and threatens violence. He takes a couple of steps in and shoots a chair at the staff with quite fiendish power. The chair hits L who collapses onto the sofa. P locks the door to the office. He roars and throws furniture at the door. Shouting that he’s going to murder us and that he’s going to stick the knife in L. Three police cars arrive and take him into custody.

Measure to prevent a repeat of the injury: For two years the staff have been pointing out that the home is not suitable for [name] and that the staffing level is too low (working alone). [Name’s] violent outbursts have become more intense. The staff have pointed this out to the manager on several occasions. We have asked for more staff, alarms and a reformulated action plan for threats and violence. Groups of staff have been replaced on several occasions as a result of excessive psychological stress and physical violence.

In the following example, the question of responsibility is once again a problem. It is unclear however whether the individual filing the report is proposing that the patient should be given care on the basis of coercive care legislation and that this would reduce the risk of externalizing behavior, or whether the patient should have been cared for at a different department with more and/or different resources. It also raises the question of how well-prepared emergency psychiatric departments are for violence. Maybe you just have to be ill enough but not too ill in order to seek care at an emergency psychiatric department?

Nurse on an emergency psychiatric ward (637)

The patient calls 911 from the corridor of the emergency psychiatric department. Carer informs him that he can’t go on calling 911. Then receives a hard punch on the right side of the face.

Measure to prevent a repeat of the injury: The right care form for patients. Not allowing this “type” of patient to “wander around departments” with no additional measures, e.g. care in accordance with the HSL [Health and Medical Services Act]. These patients are given care at different clinics/departments because nobody wants to assume responsibility.

Physical Routines and Configuration of Premises

Situationally focused, technical crime prevention has shown itself to be of significance for the prevention of violence (Welsh et al., 2013). In the retail sector, for example, cash-handling procedures and restricted opening times have shown themselves to be effective means of reducing exposure to threats and violence (Mayhew, 2004). Measures of this kind are also increasingly being tested in other sectors. The technical measures proposed in the occupational injury reports often relate to bigger or better adapted premises. The example subsequently shows that the configuration of work premises can produce strains for patients and staff.

Hospital nurse, psychiatric department (95)

Approached a worked-up and aggressive patient in order to get him to calm down and go into his room to talk. I was given a hard punch on the temple and eyebrow.

Measure to prevent a repeat of the injury: The premises where we workers are, are ridiculously small. Patients and staff are packed into spaces that are too small. Violence and threats of violence have increased as the maximum number of places has increased. The patient was medicated for the purpose of sedation. A lasting reduction of the number of places in the department.

Physical Work Environment Focused on Security

Other technical measures proposed focus on controlling who comes into and out of the workplaces, an example being emergency rooms:

Emergency room nurse (146)

Ex-patient came into the department and said that he was wearing an explosive and had the detonator in his hands. I was in his immediate presence for four hours, until the police succeeded in arresting him. Throughout this time I was convinced that the threat was real, but could not leave him because of the calming influence I was having.

Measure to prevent a repeat of the injury: Control of visitors—emergency room—at night.

The Deviance Frame

Individualization of Responsibility

Society’s attitude toward criminality in general influences the view of how work-related violence should be prevented. A study of trends in the attention focused by the trade press on measures to combat violence at work shows that there has been an increase over time in the level of demands that violence at work should be dealt with externally by means of justice system measures directed at individuals. The study also shows that this trend has occurred in parallel with a decline in the discussion of measures that involve dealing with the violence internally at the workplace where it occurs.

Unlike an occupational injury report, reporting an incident to the police can potentially involve negative consequences for a patient (Åkerström, 1997, p. 128). On the other hand, an occupational injury report can contribute to improving and developing the workplace and can serve to protect both colleagues and patients. In the following occupational injury report, it is a nurse who had been assigned to keep an anxious patient under close observation who was injured. She states that her injuries have also resulted in a report being filed with the police:

Nurse, psychiatric department (167)

The patient became angry because she didn’t get medication. The patient started to kick and scratch me on the leg, hip and hand.

Measure to prevent a repeat of the injury: “calming the patient down”, another department for aggression would suit the patient, with more staff. The injuries have been reported to the police.

Juridification

The process by which a political or ethical situation, or some other kind of situation or issue, becomes a matter for the legal system, or begins to be dealt with as a legal issue, is referred to as juridification (Garland, 2001; Mathiesen, 1985). The fact that an issue becomes juridified does not mean that everyone starts to treat the issue as a legal matter or that it is inevitable that the issue will be dealt with in this way. What it means is that a group of people at a certain point in time and in a certain context begin to approach the issue on the basis of a legal discourse. This shift need not mean that an issue is exclusively dealt with within the framework of a legal discourse, but may rather involve the legal discourse becoming more prominent in relation to other discourses.

In the following example, the police are contacted to fetch an autistic girl who did not want to go out.

Tutor, short stay home for children and youths with autism (410)

One of the children didn’t want to go out. We tried a lot of different things. After a while the child was feeling pressured and hit me on the head. I hit the other side of my head on a cupboard too. She is so angry and violent that you call the police and the child is driven to hospital.

Measure to prevent a repeat of the injury: A staff member should not be alone with the child.

The situation is resolved by someone calling the police, but the countermeasure proposed in the occupational injury report is that staff should not be alone with the child, which indicates that the police report in this case may have been filed as a way of showing the staff the seriousness of the situation and for the sake of the girl’s safety rather than because of the danger posed by the child.

Self-Defense

Nurse, psychiatric treatment home (676)

After being asked to remove his outer garments, the patient attacked his contact person. He knocked her to the floor and as she lay there she was hit and kicked repeatedly on the head and a large part of the body. The assault continued for a considerable time before the staff were able to intervene.

Measure to prevent a repeat of the injury: Course in self-defense. Personal attack alarms.

It used to be common for staff to be given training in self-defense and holding techniques in order to be able to use these when dealing with aggressive patients. One example of work of this kind is known as the Bergen model⁷ includes training in both dealing with patients and in physical interventions (Björkdahl, 2010). Measures of this kind are said to be on the way out, however (Bowie, 2002; Paterson et al., 2009), because they imply that it is the responsibility of the staff themselves to deal with the problems. This can lead to blaming the staff rather than investigating what it was that triggered the aggressive behavior of the patient/client. In the example subsequently, the incident appears to be fairly minor, but the measures proposed are extensive, which might be interpreted as an attempt to document the extent of the workload.

Carer, habilitation for the deaf and deaf–blind (342)

Staff and care recipients were sitting on the floor together, care recipient stood up, lost balance and stepped right on my foot.

Measure to prevent a repeat of the injury: Courses in self-protection and conflict-management, continuous follow-up of treatment programs with psychologist as tutor. Basic training in accidents. Continuous training in sign language. Consultations with deaf psychiatrists.

The Prevalence of Proposed Measures and Trends over Time

Which of the described frames are most common? In order to describe the situations in which the violence has occurred, this section first presents the prevalence of these situations and whether there have been changes during the period examined. The presentation then moves on to the nature of the reported violence in order to relate this to the prevalence over time of the proposed countermeasures. This is followed by a presentation of the distribution of the types of measures proposed across different areas of the health care sector.

Which Risk Situations Are Described in the Occupational Injury Reports?

It might be assumed that the types of countermeasure proposed in occupational injury reports are linked to the situations in which the violence occurred. On the basis of an interpretation of what has been written by those filing the occupational injury reports, it has been possible to identify a risk situation in three quarters of the reports included in the study. The remaining reports contained insufficient information to be categorized in this respect. The risk situations described in the occupational injury reports are working with mental illness, dementia, or neuropsychiatric disabilities. The violent incidents described in the reports often also occur in situations that involve increased intimacy, such as the changing of incontinence garments and taking care of personal hygiene, and also at meal times. According to the reports, threats and violence often occur in connection with reprimands and the communication of negative decisions. Another risk situation that is described in a number of the occupational injury reports is stress resulting from high workload. Additional risk situations that are described, although to a lesser extent are the control of unauthorized persons, being under the influence of alcohol or drugs, and the handling of valuables. These risk situations have also been documented in previous research (Piquero, Piquero, Craig, & Clipper, 2013; Viitasara, 2004; Wikman et al., 2010).

Table 3 shows an increase in the proportion of occupational injury reports that result from violence among patients in psychiatric care. In summary, the table indicates small changes in the nature of the risk situations over time, which is an important factor to bear in mind when the article moves on to interpret possible changes over time in the countermeasures proposed in the occupational injury reports.

Table 3.

Risk Situations Identified in Occupational Injury Reports 1987, 1997, and 2007.

Risk Situation	1987 (n = 423)	1997 (n = 273)	2007 (n = 231)
Mental illness^a	26	27	39
Intimate situation^b	27	24	16
Reprimand/negative decision	20	18	11
Meal times	5	6	9
Stress due to high workload	1	0	5
Under influence of alcohol or drugs	3	1	2
Control of unauthorized persons	1	1	1
Handling cash or valuables	1	0	1
Insufficient information	16	23	17
Total	100	100	100

Note. n = 927. Distribution in percentage.

^aFor example, dementia, anxiety, neuropsychiatric disabilities. ^bFor example, changing incontinence garments, taking care of personal hygiene.

Changes Over Time in the Nature of the Violence Described

Table 4 shows that the seriousness of the violence described in occupational injury reports varies over time. It is the nonminor violence that is most common throughout the period examined. The most serious violence is the second most common form reported in 1987 and minor violence is the least common form. In 2007, the relationship between these two categories of violent incidents is reversed.

Table 4.

Seriousness of the Violence Described (Minor, Nonminor, or Serious).

Seriousness of the Violence	1987 (n = 522)	1997 (n = 399)	2007 (n = 328)
Minor	18	32	29
Nonminor	50	47	48
Serious	32	22	23

Note. Proportion of occupational reports containing information on the seriousness of the violence 1987, 1997, and 2007 (n = 1,249). Values are expressed in percentage.

Since certain years may include more of the “least serious violence,” the next analysis, which focuses on change, focuses only on the nonminor and serious violence categories (see the Operationalizations section).

Which Countermeasures Are Proposed in the Occupational Injury Reports?

What measures do those who have been exposed to nonminor and serious violence propose to avoid this violence being repeated? Over half of the occupational injury reports do not include any proposed countermeasures. The proportion which include a proposed countermeasure increases over time however (1987, 30%; 1997, 48%; and 2007, 62%). Table 4 is based only on those occupational injury reports that include a proposed countermeasure, and shows how often different types of countermeasure have been proposed. Since a single report may include several measures, the columns sum to over 100%. The year with the highest column total (2007) is thus the year in which the largest number of measures was proposed per occupational injury report. There is no major increase in the number of measures proposed on the occupational injury reports and the majority (59%) only include one proposed measure.

Table 5 shows that the normalization frame, with its countermeasures focused on improving the psychosocial work environment, is by far the most common. The three measures that go to make up this overarching category are guidance and crisis management, more resources, and sharing knowledge. At the beginning of the period examined, resources is the most commonly proposed measure of the three, whereas guidance and crisis management is the most common at the end of the period. The number of proposals relating to guidance and crisis management increases more than the number of proposals relating to any of the other countermeasures, with guidance and crisis management being proposed in no more than 1 in 10 occupational injury reports in 1987, and in one quarter of the reports filed in 2007.

Table 5.

Distribution of Countermeasure Frames and Countermeasures in Occupational Injury Reports Describing Nonminor and Serious Violence for the Years 1987, 1997, and 2007.

Frame	Countermeasures	1987 (n = 128)	1997 (n = 132)	2007 (n = 144)
Normalization	Coaching sessions	53	75	77
	More resources	74	55	46
	Sharing knowledge	9	25	24
Control	Diagnosis/medication	38	39	32
	Physical routines	25	17	23
	Physical security	19	11	16
Deviance	Self-defense training	5	12	9
	Police/offence report	0	0	2
	Active monitoring	4	6	2
	Sum (%; more than one countermeasure may be proposed)	161	161	172

Note. n = 404. Values are expressed in percentage.

The control frame, with measures focused on diagnosis/medication is found relatively frequently in the occupational injury reports. Technical solutions in the form of measures focused on adapting the physical work environment to the clients, for example, adapting the premises to the nature of the work and various technical solutions, also occur equally frequently throughout the period examined. The deviance frame, with typical measures involving filing police reports, active monitoring, and self-defense, is very rarely advocated by health care-sector employees.

Are the Same Measures Proposed for Different Types of Violence?

The health care sector has a range of different client groups and the care-sector staff may therefore be expected to have a range of strategies for dealing with violence. The proportion of occupational injury reports that include proposals for psychosocial countermeasures is on the increase not only in the material as a whole but also in relation to each type of violence examined (minor, nonminor, and serious) and across different areas of the health care sector (e.g., psychiatric care, geriatric care, and care provision for the disabled).

It can be seen from Table 6 that the normalization frame is dominant irrespective of the client group. At the same time, there are a number of small differences in the patterns found across different areas of the health care sector. The most prominent of these is that one in five of the measures proposed within the addiction care sector are located within the deviance frame, as against 1 in 20 of the measures proposed in the context of geriatric care.

Table 6.

Distribution of the Frames Across Different Health Care Sectors.

	Children	Psychiatry	Addiction	Health Care Centers	Geriatric	Disability	Total
Normalization	67	47	42	52	60	63	56
Control	21	44	39	39	36	27	35
Deviance	12	9	19	9	4	9	9
Total (n)	100 (42)	100 (231)	100 (31)	100 (95)	100 (154)	100 (219)	100 (852)

Note. Proportion of proposed countermeasures (ISA 1987, 1997, and 2007). ISA = The Swedish Information System on Occupational Accidents and Work-related diseases.

Values are expressed in percentage.

Discussion

Occupational injury reports capture a portion of the most serious violence to which health care staff is exposed at work. If the societal tendency toward juridification and the responsibilization of the individual (Garland, 2001, p. 124, Schindeler, 2013) has influenced the way care staff approach the issue of threats and violence at work, then this should be visible in the occupational injury reports examined here, since it is first and foremost the most serious violence that requires interventions from external actors such as the police and the justice system. There is nothing in the material examined in the current study, however, that would suggest such an influence.

The staffs exposed to violence in the health care sector most often propose countermeasures from within the normalization frame with a focus on improving the psychosocial work environment. Irrespective of the type of risk situation that preceded the violent incident, the staff would prefer to see the problems resolved internally at the workplace, for example, by sitting down and talking through what happened with other staff members in the form of coaching sessions or debriefing. Time and again the staffs ask for resources in the form of the time and staff required to do the work. Looking to the control frame, the violence problem is sometimes “resolved” by transferring responsibility to someone else, either by defining the patient as being too ill to be given care or as being responsible for his or her actions, despite the fact that the violence of these patients and clients is often only a case of them acting in accordance with their diagnosis or illness. Proposals involving countermeasures from the deviance frame in the form of repressive interventions on the part of the police and the justice system are uncommon.

The finding that those exposed to violence do not place blame on particular individuals is in line with the results reported by several other studies (Jones et al., 2011). It is not a question of a small number of “bad apples” or “one-offs” who are disrupting the work of the health care sector, but instead the violence is perceived to occur as a result of the fact that people working in certain areas of health care, irrespective of individual characteristics such as social class, age, or sex, are at higher risk of being confronted with these clients as a result of the “nature of the work.” The reason that the staffs do not turn to the police and the state control apparatus is probably that the care sector is precisely what its name implies, a system of care: If the staff turn to the police and the justice system, the patients cease to be “patients” and they themselves cease to be “caring” (cf. Åkerström, 2002).

A difference between the findings of this study and those of Åkerström’s (1993) study of threats and violence against health care staff, in which approximately one quarter of employees stated a desire for training in holding techniques and self-defense, very few of the staff in the current study propose measures of this kind. This difference may be due to the fact that the focus in this study has been on those exposed to serious violence, whereas Åkerström’s study was based on a sample of health care employees and not specific individuals who had been exposed to violence and to the fact that self-defense training was one of the predetermined response alternatives⁸ included in Åkerström’s study.

While there is little evidence as to the efficacy of incident reporting systems in preventing and minimizing workplace violence, there is a good deal of evidence that poor reporting practices are the norm rather than the exception in medical and other health care settings (Hills & Joyce, 2013). One objection that may be raised in relation to the countermeasures proposed by staff is that they may propose certain measures because asking for more resources is an easy option. In this respect, however, there have been changes over time. Whereas previously staff did indeed primarily propose improved resources, today staff more often states a desire for crises counseling/debriefing (see Carlsson & Wennerström, 2010; Paterson et al., 1999). We cannot tell from the occupational injury reports whether the staffs want more time to be made available for internal coaching sessions or whether they want more frequent or a different form of coaching sessions. These measures do however imply that the perpetrator of the violence is viewed as a victim of circumstances that he or she has no control over.

It is notable that the move toward juridification and responsibilization that had characterized both developments in the crime policy arena at large and the debate in the trade press, does not manifest itself among those who work within the health care sector, even if the focus is restricted to those who have themselves been exposed to violence and threats at work. This finding may be interpreted as indicating that there may be a resistance to responding positively to proposed measures within health care organizations—among those responsible for health care operations and managers who have the authority to implement such measures. This situation is not unique to the health care sector, however, but is also found in other sectors where accidents occur (Lundberg, Hollnagel, Rollenhagen, & Rankin, 2011) and may be due to the fact that the available countermeasures influence the way the problem is understood. People’s firsthand knowledge of crime often contradicts the pictures painted by politicians and the media. The individual who is exposed to violence analyses the situation on the basis of a knowledge base that differs from that of those located further away from the sequence of events leading up to the incident and who have to develop their understanding of such incidents retrospectively.

Several studies on violence at work and the psychosocial work environment have found correlations between a sense of having little control, high demands, stress, and a lack social support (Wikman et al., 2010). One of the dangers associated with the failure to respond positively to proposals for countermeasures from health care staff is that this may lead to an intensification of the sense of facing excessive demands and having little control, which may in turn produce more stress and thus also result in more exposure to violence.

While there is little evidence as to the efficacy of incident reporting systems in preventing and minimizing workplace violence, there is a good deal of evidence that poor reporting practices are the norm rather than the exception in medical and other health care settings (Hills & Joyce, 2013). One study limitation that may be raised in relation to the countermeasures proposed by staff is that they may propose certain measures because asking for more resources is an easy option. In this respect, however, there have been changes over time. Whereas previously staff did indeed primarily propose improved resources, today staff more often states a desire for coaching sessions. We cannot tell from the occupational injury reports whether the staffs want more time to be made available for internal coaching sessions or whether they want more frequent or a different form of coaching sessions. These measures do however imply that the perpetrator of the violence is viewed as a victim of circumstances that he or she has no control over.

There are many questions that cannot be addressed in detail via analysis of injury reports such as these. My division of the type of violence injuries into minor, nonminor, and severe does not take into account the psychological injuries that may have occurred and that aspect could be further developed in another study. Knowledge about violence prevention appears to exist at workplaces themselves, but these are not given the resources they need to resolve the problem. A more extensive and robust evidence base is required to enable informed decision making on reducing the likelihood of violence as well as psychosocial injuries in health care settings.

This article may serve as providing an empirical example of a paradox. Although the objective of crime prevention today has considerable legitimacy and is also furnished with relatively substantial resources, we are at risk of completely failing to prevent violence as a result of having an overly simplistic understanding of the problem. To the extent that we lack both an analysis of the negative effects of social change on the prevalence of violence at work and any interest in using measures other than legal strategies focused on individuals, we risk finding ourselves in a situation where measures to combat threats and violence at work are restricted to a focus on personal interactions between staff and clients. What we are failing to address are the underlying factors that serve to structure the framework in which these interactions take place. A security research perspective could be of assistance in the systematic charting of the correlations between exposure to violence at work and the situational conditions that may contribute to producing incidents of threats and violence. This might involve studying the governance and management of work tasks, resources, working conditions, competence, reporting, and analyses of security activities.

Footnotes

Acknowledgment

I would like to thank David Shannon for translating the text.

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author disclosed receipt of the following financial support for the research and/or authorship of this article: The article was written in the framework of the Swedish research Council for Health, working life, and welfare (FORTE) —financed project “Violence and Threat Risk Assessment in three Government Agencies.”

Notes

References

Åkerström

(1993). Våld och hot i sjukvårdsarbeten. Dokumentation av en riksrepresentativ intervjustudie. Lund, Sweden: Network for Research in Criminology and Deviant Behaviour at Lund University.

Åkerström

(1997). Att anmäla våld—en praktik inom lokala kulturer. Nordisk Tidsskrift för Kriminalvidenskab, 84, 125–132.

Åkerström

(2002). Slaps, punches, pinches—But not violence: Boundary-work in nursing homes for the elderly. Symbolic Interaction, 25, 515–536.

Arbetsmiljöverket. (2011). Arbetsmiljöstatistik. Arbetsskador 2010. Report 2011:1. Stockholm, Sweden: Author.

Balvig. (1979–1980). Samfundsplanlægning og kriminalitetsforebyggene tiltag. Nordisk Tidsskrift for Kriminalvidenskab, 8, 96–111.

Benford

R. D.

Snow

D. A.

(2000). Framing processes and social movements: An overview and assessment. Annual Review of Sociology, 26, 611–639.

Björkdahl

(2010). Violence prevention and management in acute psychiatric care: Aspects of nursing practice. Stockholm, Sweden: Karolinska universitetssjukhuset.

Brå. (2009). Grövre våld i skolan. Report 2009:6. Stockholm, Sweden: Brottsförebyggande rådet.

Braverman

(2002). The prevention of violence affecting workers: A systems perspective. In Gill

Fisher

Bowie

, (Eds.), Violence at work: Causes, patterns and prevention (pp. 114–131). Cullompton, England: Willan.

10.

Bowie

(2002). Defining violence at work: A new typology. In Gill

Fisher

Bowie

(Eds.), Violence at work: Causes, patterns and prevention (pp. 1–20). Cullompton, England: Willan.

11.

Bowie

(2010). An emerging awareness of the role organizational culture and management style can play in triggering workplace violence. Workplace Violence in the Mental and General Healthcare Setting, 1, 43–58.

12.

Bryman

(2012). Social research methods. Oxford, England: Oxford University Press.

13.

Carlsson

Wennerström

U.-B.

(2010). Kartläggning och genomlysning av stödet till brottsutsatta personer i Göteborg. Göteborg, Sweden: Institutionen för socialt arbete/Sociologiska institutionen, Göteborgs Universitet.

14.

Chappell

Di Martino

(2006). Violence at work. Genevé, Switzerland: ILO.

15.

Christie

(1999). Forebyggelse. Nordisk Tidsskrift for Kriminalvidenskab, 84, 5–15.

16.

Estrada

Nilsson

Jerre

Wikman

(2010). Violence at work—The emergence of a social problem. Journal of Scandinavian Studies in Criminology and Crime prevention, 11, 46–65.

17.

European Agency for Safety and Health at Work. (2011). Occupational safety and health culture assessment—A review of main approaches and selected tools. Luxemburg, IA: Publications Office of the European Union.

18.

Garland

(2001). The culture of control: Crime and social order in contemporary society. Oxford, England: Oxford University Press.

19.

Giertsen

(1994). Forebygging av kriminalitet: Altfor bred tog altfor smalt. Nordisk Tidsskrift for Kriminalvidenskab, 3, 295–313.

20.

Gieryn

T. F.

(1983). Boundary-work and the demarcation of science from non-science: Strains and interests in professional ideologies of scientists. American Sociological Review, 48, 781–795.

21.

Goffman

(1974). Frame analysis: An essay on the organization of experience. New York, NY: Harper & Row.

22.

Gustafsson

Szebehely

(2005). Arbetsvillkor och styrning: äldreomsorgens hierarki—en enkätstudie bland personal och politiker. Edsbruk, Sweden: Akademitryck.

23.

Hills

Joyce

(2013). A review of research on the prevalence, antecedents, consequences and prevention of workplace aggression in clinical medical practice. Aggression and Violent Behaviour, 18, 554–569.

24.

Hydén

(2004). Hur får vi en bättre arbetsmiljö? Om strategier i arbetsmiljölagstiftningen över tid och vad vi kan lära av det. In Johansson

Frick

Johansson

(Eds.), Framtidens arbetsmiljö—och tillsynsarbete (pp. 126–151). Lund, CA: Studentlitteratur.

25.

International Labor Office/World Health Organization/International Council of Nurses/Public Services International. (2000). Workplace violence in the health care sector: Country Case study—Questionnaire. Geneva, Switzerland: ILO.

26.

Jareborg

Zila

(2007). Straffrättens påföljdslära. Stockholm, Sweden: Nordstedts Juridik.

27.

Jones

Robinson

Fevre

Lewis

(2011). Workplace assaults in Britain. Understanding the influence of individual and workplace characteristics. British Journal of Criminology, 51, 159–178.

28.

Larsson

(2004). Förutsättningar för kommunernas arbetsmiljöarbete på 2000-talet. In Johansson

Frick

Johansson

(Eds.), Framtidens arbetsmiljö- och tillsynsarbete (pp. 184–221). Lund, CA: Studentlitteratur.

29.

Lundberg

Hollnagel

Rollenhagen

Rankin

(2011). Strategies for dealing with resistance to recommendations from accident investigation. Accident Analysis & Prevention, 45, 455–467.

30.

Mathiesen

(1985). Rätten i samhället. Göteborg, Sweden: Bokförlaget Korpen.

31.

Mayhew

(2004). Occupational violence in industrialized countries. In Gill

M. L.

Fisher

B. S.

Bowie

(Eds.), Violence at work, causes, patterns and prevention (pp. 21–40). London, England: Willan.

32.

Paterson

Leadbetter

Bowie

(1999). Supporting nursing staff exposed to violence at work. International Journal of Nursing Studies, 36, 479–486.

33.

Paterson

Leadbetter

Miller

Bowie

(2009). Re-framing the problem of workplace violence directed towards nurses in mental health services in the UK: A work in progress. International Journal of Social Psychiatry, 56, 310–320.

34.

Petersson

(2012). Utbrändhet som arbetsskada och motståndsberättelse. Sociologisk forskning, 49, 211–226.

35.

Piquero

N. L.

Piquero

A. R.

Craig

J. M.

Clipper

S. J.

(2013). Assessing research on workplace violence, 2000–2012. Aggression and Violent Behavior, 18, 383–394.

36.

Sahlin

(2000). Brottsprevention som begrepp och samhällsfenomen. Lund, CA: Arkiv.

37.

Schindeler

(2013). Workplace violence: Extending the boundaries of criminology. Theoretical Criminology, 1–15. Published online before print December 5, 2013. doi:10.1177/1362480613511980

38.

Viitasara

(2004). Violence in caring: Risk factors, outcomes and support. Stockholm, Sweden: Arbetslivsinstitutet.

39.

Waddington

Badger

Bull

(2005). Appraising the inclusive definition of workplace violence. British Journal of Criminology, 45, 141–164.

40.

Welsh

Bader

Evans

S. E.

(2013). Situational variables related to aggression in institutional settings. Aggression and Violent Behavior, 18, 792–796.

41.

Wikman

(2012). Våld i arbetslivet. Utveckling, uppmärksamhet och åtgärder. Stockholm, Sweden: Stockholms universitet, Kriminologiska institutionen.

42.

Wikman

Estrada

Nilsson

(2010). Hot och våld i arbetslivet. En kriminologisk kunskapsöversikt. Report 2010:4. Stockholm, Sweden: Arbetsmiljöverket.

43.

Work Environment Act, 1160 (1977).