Abstract
Psychological aggression perpetrated by customers, coworkers, and supervisors is a behavior frequently experienced in the workplace with negative consequences for an individual’s health. The aim of this systematic review was to examine the personal outcomes of overt workplace psychological aggression and summarize empirical evidence on how to prevent and reduce its effects. A search on PubMed, Scopus, ProQuest, and Web of Science electronic databases was made. Data were obtained from 30 studies (26 cross-sectional, 3 longitudinal, 1 quasi-experimental) representing 20,683 employees. Longitudinal studies indicated that workplace psychological aggression is significantly associated with musculoskeletal injury and psychological strains (anxiety, depression, anger) over time. Research also suggests that psychological aggression at work predicts fear, emotional exhaustion, depersonalization, personal accomplishment, and physical symptoms, for example. Additionally, the information collected suggested that support from coworkers, supervisors, and management; informational support; political skills; job resources; and confidence to prevent and respond to aggression moderate significantly the impact of workplace psychological aggression on personal and organizational outcomes. Perceptions of a violence-prevention climate, aggression-preventive supervisor behavior, and aggression-preventive employee effort are also significantly associated with a reduction in psychological aggression at work and strains. In conclusion, this review allows us to understand the effects of the psychological aggression by identifying the preventive strategies that could be adopted by managers, supervisors, or leaders to deal with it and promote individual’s health in the workplace.
Introduction
Subjacent to this review are the importance of work and its central role in the lives of the general population and its potential to promote and sustain mental health. Interest in work and work transitions across the lifespan has grown, in part explained by the evidence that is positively associated with behavioral, emotional, and physical aspects of an individual’s health and their impact on other domains of people’s lives (American Psychological Association [APA], 2016; Blustein, 2008, 2013; see also Duffy, Blustein, Diemer, & Autin, 2016).
In this systematic review, our focus is the relationship between workers’ experience of psychological aggression and strains, and resources to prevent and minimize their effects.
There has been a wide debate in research on conceptual and operational issues of various forms of human aggression that occur at work, and we can find a diversity of definitions and related terms (e.g., Aquino & Thau, 2009; Barling, Dupré, & Kelloway, 2009; Bowling & Beehr, 2006; Bowling, Camus, & Blackmore, 2015; European Agency for Safety and Health at Work, 2010; International Labour Organization [ILO], 2013). In this respect, the aggression and organizational literature also recognizes distinctive constructs, with theoretical differences. For example, bullying, workplace aggression (physical and psychological aggression), incivility, and abusive supervision differ in terms of intensity and intent to harm (Nielsen, Matthiesen, & Einarsen, 2010; Schat & Kelloway, 2005; Tepper & Henle, 2011; Yang, Caughlin, Gazica, Truxillo, & Spector, 2014; see also Nielsen & Einarsen, 2018).The World Health Organization (WHO, 2019) argues that workplace violence involves both physical and nonphysical violence in work-related circumstances, but it does not clearly distinguish the different constructs mentioned above (see also ILO, 2013).
Thus, in this study, we offer an examination of workplace psychological aggression. Different terminology (as we find in other specific constructs—cf. Tepper, Simon, & Park, 2017) has been used by researchers to designate this form of workplace aggression, such as nonphysical aggression (e.g., Yang et al., 2014), verbal aggression (e.g., Kessler, Spector, Chang, & Parr, 2008), psychological aggression (e.g., Dionisi, Barling, & Dupré, 2012; Schat & Frone, 2011), and nonphysical violence (e.g., Spector, Zhou, & Che, 2014). In this review, we apply the term workplace psychological aggression, employed by Kelloway and colleagues (e.g., Barling et al., 2009; Schat & Kelloway, 2003), in line with the distinction between verbal aggression and psychological aggression made by Straus, Hamby, Boney-McCoy, and Sugarman (1996). We adopt Schat and Kelloway’s (2005) perspective and definition of workplace aggression as “a behavior that is enacted by an individual(s) within or outside an organization, intended to physically or psychologically harm a worker or workers, and occurs in a work-related context.” (p. 191). This is consistent with our aims for the following reasons.
First, it is difficult to examine the effects of physical violence independently of psychological aggression because those who experience physical violence also experience psychological aggression. Only a few employees reported physical violence without experiencing psychologically aggressive behaviors. There is also consensus that not all aggressive behaviors are violent (Schat & Frone, 2011; Schat & Kelloway, 2005). Physically violent and psychologically aggressive behaviors experienced at work such as being hit and yelled at can co-occur, but they are different forms of aggression if we consider their immediate effects: physical and psychological harm (Schat & Frone, 2011). Evidence shows that although related, physical violence and psychological aggression are different constructs and empirically distinguishable. This approach is important because it leads us to focus on a range of psychological aggressive behaviors toward employees (e.g., yelling) that occur more frequently than physical violence and tend to be less visible (Barling et al., 2009; Schat & Frone, 2011).
Second, as stated by Schat and Kelloway (2005), this definition differs from other related constructs when we consider the following: (i) the frequency or duration of aggressive behaviors. The experience of negative acts is not necessarily repeated and prolonged in time as we found in definitions of bullying or harassment (cf. Nielsen, Glasø, & Einarsen, 2017; Nielsen et al., 2010; see also Nielsen & Einarsen, 2018); (ii) the target is a worker or workers. This feature distinguishes this form of aggressive behavior from organizational counterproductive behaviors such as theft, sabotage; (iii) the source of aggression can be from within (colleague) or outside (customers) the organization. The perpetrator is not limited to any one source, as we found in abusive supervision (i.e., supervisors; Tepper et al., 2017), for example.
In addition, contrary to what we found in other related constructs, such as discrimination or sexual harassment, acts of aggressive behaviors toward employees are not necessarily motivated by race or gender (Hershcovis & Barling, 2010a).
Third, the perspective of Schat and Kelloway (2005) goes further by considering experience of vicarious events. Evidence suggests that one need not experience aggressive behaviors directly to be affected (Barling, 1996). In fact, like direct aggression, workplace vicarious aggression is also associated with negative outcomes (Dupré, Dawe, & Barling, 2014; Schat & Kelloway, 2003).
Consistent with previous research, the individual’s perceptions of aggressive behaviors can lead to psychological stress, which in turn generates psychological and physical strains (Barling, 1996; Schat & Kelloway, 2005). Supported by the traditional work stress framework (stressor→stress→strain) and Bandura’s (1977) social learning theory, this model has been widely used by researchers to examine the negative impact of workplace aggression on personal and organizational outcomes. For example, direct and vicarious workplace aggression by customers affects the perceived risk of aggression, which in turn influences physical health, mental health, affective commitment, and turnover intentions (Dupré et al., 2014). The negative consequences of the experience of workplace aggression can go further and affect other domains of an individual’s life. Coworker workplace aggression is linked to psychological detachment, which in turn is associated with work–family conflict (Demsky, Ellis, & Fritz, 2014).
Other outcomes were demonstrated in Hershcovis and Barling’s (2010b) meta-analyses. Combining a variety of constructs (e.g., incivility, bullying, mistreatment) into a broad aggression category, the study showed that (i) supervisor aggression has stronger negative effects than coworker aggression on organizational outcomes (e.g., turnover intentions); (ii) coworker aggression has stronger negative effects than outsider aggression on organizational outcomes (e.g., job satisfaction, commitment); and (iii) there was no significant difference between supervisor, coworker, and outsider aggression in relation to emotional exhaustion and depression.
Workplace aggression is also associated, for example, with shame, low self-esteem, frustration, job stress, posttraumatic stress, and somatic symptoms (Aquino & Thau, 2009; Bowling et al., 2015).
Another critical issue in this area has been to identify specific ways to prevent workplace aggression. Research on the nature of workplace aggression antecedents has found that organizational and contextual factors (e.g., role ambiguity, organizational injustice) are associated with workplace aggression, so eliminating them is crucial for organizations (Dupré et al., 2014). On the other hand, research focus on the consequences of workplace aggression is particularly interesting regarding moderator variables, such as perceived control and social support, that may reduce the effect of workplace aggression on outcomes (Barling, 1996; Schat & Kelloway, 2005).
This review is pertinent due to the prevalence of psychological aggression in the workplace compared to other distinct constructs related with aggressive behaviors at work (Nielsen et al., 2010; Tepper et al., 2017). The review by Spector et al. (2014) of the literature on violence toward nurses shows that rates of exposure to nonphysical violence were significantly higher than those of bullying, physical violence, and sexual harassment (i.e., 66.9%, 39.7%, 36.4%, and 25%, respectively). According to these authors, patients and family/friends perpetrated most nonphysical violence, with 53.9% and 47.3%, respectively.
Unlike past reviews, the novelty of our study lies in examining workplace psychological aggression as a specific construct. The workplace aggression variable used by Yang et al. (2014) includes overt physical and nonphysical behaviors. Hershcovis and Barling (2010b) include in one broad workplace aggression variable measures of bullying, incivility, mobbing, and petty tyranny, for example. Dudenhöffer and Dormann’s (2015) and Tepper’s reviews (Tepper, 2007; Tepper et al., 2017) are based on one focal construct and instrument: customer-related social stressors (Dormann & Zapf, 2004) and abusive supervision (Tepper, 2000), respectively. In line with Nielsen and Einarsen (2018), we also believe that workplace psychological aggression should be considered from another angle for greater understanding of this phenomenon, in order to develop better programs and guidelines to prevent it.
Aims
Our aims are to examine the personal outcomes of workplace psychological aggression and identify a set of variables that buffer those outcomes. Specifically, we focus on targets’ perceptions of overt psychologically aggressive behaviors with the intent of inflicting harm, perpetrated by customers, coworkers, and supervisors and experienced directly (i.e., employees being yelled, shouted or sworn at) or vicariously (i.e., employees that have seen or heard about a colleague being yelled or shouted at), rather than ambiguous behaviors (e.g., incivility or social undermining) or employees’ aggressive behaviors toward individuals or the organization (perpetrator perspective). We examine the effects of this experience on personal outcomes. That is not to say that our construct is not associated with negative organizational outcomes, but these strains are not our primary interest here. In addition, we summarize direct associations and moderating effects of organizational and individual third variables (MacKinnon, 2011) that will potentially modify positively the relation between psychological aggression and personal outcomes. In doing so, we believe that effective ways to design future preventive interventions are identified.
Methods
Protocol and registration
Details of the protocol for this systematic review were registered on PROSPERO (CRD42017064400) and is available in full on the NIHR HTA program website (http://www.crd.york.ac.uk/PROSPERO).
Eligibility criteria
Based on the literature on workplace aggression (e.g., Barling, 1996; Schat & Kelloway, 2005), inclusion criteria were applied to ensure transparency and similar interpretation in the studies selected.
Empirical records in English, published between 2000 and 2017, were retrieved: (i) where workplace psychological aggression is defined; (ii) analyzing workplace psychological aggression from the target perspective; and (iii) reporting association between workplace psychological aggression and personal outcome variables. Adult (18 years or older), full- or part-time employees, whatever the occupation and the work setting, were included.
All studies selected for inclusion in this systematic review were published in peer-review journals, so reviewed by experts according to rigorous standards of scientific research.
Literature search
A search of the literature was made between April and May 2017. First, we searched the electronic bibliographic databases PubMed, Scopus, ProQuest, and Web of Science (Science and Social Science Citation Index). The search terms used to identify relevant studies included, for example, “workplace”/“work”/“occupational” AND “psychological aggression”/“verbal aggression”/“verbal violence”/“nonphysical aggression”/“nonphysical violence.” Second, a search of APA and affiliated journals was carried out on Google Scholar. Finally, we also performed a manual search of the reference list included in meta-analysis and review articles. Any personal outcome associated with workplace psychological aggression was considered.
The selection process was conducted in accordance with a four-phase Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram (Moher, Liberati, Tetzlaff, & Altman, 2009). Based on the title and/or abstract information, two researchers independently identified records through database searching, removed duplicate records, screened, and excluded records. Two reviewers assessed full-text articles for eligibility with those not meeting the criteria being excluded. A third reviewer checked a random sample of included articles to verify if the criteria were being met consistently.
Those studies selected for inclusion also were subject to critical appraisal by two reviewers independently based on the Joanna Briggs Institute tools (Moola et al., 2017). Disagreements related to quality assessment were successively resolved by discussion in the review team. The results of this appraisal highlight that 15 out of 30 studies reported exposure measure reliability and 19 out of 30 studies identified both demographic characteristics (e.g., gender, age) and work-related factors (e.g., tenure, hours worked, work site) as control variables and strategies to deal with them. None of the studies was excluded.
Data items
Relevant information for each sample was gathered within the following criteria: (i) exposure measure is identified as in the original study; (ii) tenure (years) was not reported due to insufficient data; (iii) source of aggression was categorized as customer (e.g., client, user), coworker (e.g., colleagues), supervisor (e.g., superiors); (iv) online survey data collection includes questionnaires distributed by e-mail; (v) outcome measures are identified by their authors; (vi) personal outcomes are reported regardless of their statistical significance; and (vii) significant third variables (individual or organizational) related with resources and strategies to prevent workplace psychological aggression are reported.
Results
Study selection
A flow diagram (see Figure 1) shows the numbers of records screened, removed, assessed for eligibility, and included in this review.

Studies screened, assessed for eligibility, and included in this review, with reasons for exclusions at each stage.
A total of 2,845 records were identified, 263 were screened. Of these, 216 were excluded, for the following reasons: (i) 28% of studies focus on other constructs (e.g., bullying—Balducci, Cecchin, Fraccaroli, & Schaufeli, 2012; physical violence—Mueller & Tschan, 2011) (ii) or examine the exposure measure from both the target and perpetrator perspective (e.g., interpersonal conflict—Demsky et al., 2014; interpersonal and organizational counterproductive work behavior—Meier & Spector, 2013), and (iii) 55% did not report associations between workplace psychological aggression and personal outcomes. Of the 47 full texts assessed for eligibility, 14 were excluded due to (i) exposure measure (n = 7), for example, Dupré et al. (2014) assessed direct and vicarious workplace aggression of a psychological and physical nature by customers, but all items of measurement were averaged to form one index; Nixon and Spector’s (2015) measure included verbal aggression, intimidation, social exclusion, undermining, rude behavior, interpersonal conflict, and physical aggression; (ii) no report of associations between workplace psychological aggression and outcomes (n = 7). Thus, 30 studies were included in this systematic review.
Demographic and study characteristics
More than half the studies (n = 16) were conducted in America, eight in Europe, four in Asia, and one in Africa. One study did not report the country (Dionisi et al., 2012).
Sample size varied from 118 (Itzhaki et al., 2015) to 3,471 participants (Dhaini et al., 2015). Of the total number of 20,683 participants, 10,579 worked in health care in a variety of work settings.
Twenty-six studies have a cross-sectional design, and one has a quasi-experimental design. Three studies are longitudinal. Different data collection techniques were used: traditional paper-and-pencil (n = 17), online method (n = 8), telephone survey (n = 2), telephone survey and online (n = 1), and hardcopy and online technique (n = 2).
Exposure measure
Great heterogeneity was found in the measures used to assess workplace psychological aggression. Frequency of workplace psychological aggression was measured by (i) a single item (e.g., Itzhaki et al., 2015), (ii) items adapted from other measures (e.g., Da Silva et al., 2015), (iii) continuous measures that aggregated specific workplace psychological aggression indicators (e.g., Pacheco, Cunha, & Duarte, 2016), or (iv) newly developed instruments (e.g., Dormann & Zapf, 2004; Table 1).
Study characteristics and workplace psychological aggression measures.
WHO: World Health Organization; VAW: Violence Against Women.
All the items with the symbol (˗) were not reported.
The authors have distinguished different sources.
Gender: Cases with missing data were deleted.
Dionisi et al. (2012)—Authors note: “ . . . no internal consistency data are provided.”
Itzhaki et al. (2015)—Mental Health Hospital.
Magnavita (2013)—Infectious Diseases Hospital.
Spector et al. (2007)—Veteran’s Health Administration Hospital.
The content of the exposure measures is focused on overt psychological behaviors such as “been threatened,” “yelled at,” “a door abruptly shut,” “insulted,” “something spiteful was said,” “treated rudely/with disrespect,” “interrupted while you were speaking/working,” “judged or criticized unfairly,” “address me defiantly,” “customers always complaining,” and “argue all the time” (e.g., Barling, Rogers, & Kelloway, 2001; Chang, Eatough, Spector, & Kessler, 2012; Chang & Lyons, 2012; Dionisi et al., 2012; Dormann & Zapf, 2004; Grandey, Kern, & Frone, 2007; Hanson, Perrin, Moss, Laharnar, & Glass, 2015; Kessler et al., 2008; LeBlanc & Kelloway, 2002; Llor-Esteban, Sánchez-Muñoz, Ruiz-Hernández, & Jiménez-Barbero, 2017). Other studies reported the content of the exposure measure broadly as verbal aggression and insults (Da Silva et al., 2015; Dhaini et al., 2015; Itzhaki et al., 2015; Jaradat et al., 2016; Jung et al., 2015; Magnavita, 2013).
Of the included studies, 15 provide exposure measure reliability, showing adequate internal consistency (coefficient alpha) with values ranging from α = .72 (e.g., Dormann & Zapf, 2004) to α = .93 (Pacheco et al., 2016).
Concerning the source of psychological aggression at work: (i) nine studies did not identify the source (e.g., Schat & Frone, 2011); (ii) in three studies, although the source is mentioned, subsequent analyses do not specify it (e.g., Spector, Yang, & Zhou, 2015); (iii) 13 studies addressed workplace psychological aggression perpetrated by customers; and (iv) one study (Dionisi et al., 2012) examined supervisor psychological aggression and another (Yang & Caughlin, 2017) coworker psychological aggression.
Three studies (Chang & Lyons, 2012; Grandey et al., 2007; LeBlanc & Kelloway, 2002) distinguished different sources of workplace psychological aggression. In doing so, differences associated with outcomes were also examined.
A specific period in which workplace psychological aggression occurred was reported in 22 studies varying considerably from 1 day (Grandey, Dickter, & Sin, 2004) to the past year/entire career (Itzhaki et al., 2015).
Outcomes
Consistent with our aim regarding the personal outcomes of workplace psychological aggression, a variety of outcomes were taken from studies. We found a lack of consistency in their measurement; for example, anxiety was examined in nine studies through measures such as Emotional Strain Scale, Brief Symptom Inventory 18, General Health Questionnaire, and Mood Scale.
In addition, this review has examined significant variables associated with prevention and reduction of those outcomes. Of the total of 30 studies, 14 examined a set of related individual and organizational variables. Of these studies, 10/14 studies were conducted in the USA. Moderating variables in the research design were included in 5/14 and 9/14 studies examined associations between third variables and workplace psychological aggression and strains.
Main findings
Workplace psychological aggression outcomes
In this review, the correlation between workplace psychological aggression and the outcome was retained. For example, Dormann and Zapf’s (2004) exposure measure (customer-related social stressors) has four dimensions. For our purposes, only the association between the factor of “customer verbal aggression” (as it reflects psychological aggression by customers) and the outcome was considered.
Three longitudinal studies conducted in the health-care sector show that exposure to psychological aggression at work is related with mental and physical complaints over time. Workplace psychological aggression affects anxiety and depression in baseline assessment after 12 months of exposure (Magnavita, 2013) and is significantly correlated with anger and depression 6 months and 12 months after nurses begin work (Spector et al., 2015). Zhou, Yang, and Spector (2015) found that workplace psychological aggression is a significant predictor of anger and musculoskeletal injury 12 months after nurses begin work (Table 2).
Correlations and regression coefficients between workplace psychological aggression and personal outcomes.
OR: odds ratio; CI: confidence interval.
Items with the symbol (˗) were not reported in the original study.
aItems developed by study authors.
bLevel of significance not reported.
*p < .05. **p < .01. ***p < .001.
Customer psychological aggression is significantly associated with emotional exhaustion, depersonalization and personal accomplishment (Dhaini et al., 2015; Dormann & Zapf, 2004; Grandey et al., 2004; Karatepe & Nkendong, 2014; Karatepe, Yorganci, & Haktanir, 2009; Li & Zhou, 2013; Llor-Esteban et al., 2017; Viotti, Gilardi, Guglielmetti, & Converso, 2015; Winstanley & Whittington, 2002), fear, diminished psychological well-being, physical symptoms (LeBlanc & Kelloway, 2002; Pacheco et al., 2016), back pain and joint pain (Dhaini et al., 2015) poor life satisfaction (Itzhaki et al., 2015), social dysfunction (Llor-Esteban et al., 2017), sleep problems, and stress (Hanson et al., 2015).
A few studies explored workplace psychological aggression perpetrated by supervisors and coworkers. The experience of supervisor psychological aggression is negatively associated with psychological well-being (Dionisi et al., 2012), and coworker psychological aggression is positively correlated with emotional exhaustion and physical symptoms (Yang & Caughlin, 2017).
Differences associated with psychological aggression perpetrated from different sources and outcomes were examined. Chang and Lyons (2012) discriminated four types of sources and compared associations between psychological aggression from those different perpetrators and outcomes. Their results suggested that psychological aggression by all four perpetrators (customer, coworkers, supervisor, and significant others) is positively associated with increased anger, anxiety, and depression. Grandey et al. (2007) found that all three perpetrators of psychological aggression, inside (coworkers and supervisor) and outside (customer), were significant predictors of job-related emotional exhaustion. LeBlanc and Kelloway (2002) examined employees’ experience of psychological aggression from two sources. The results suggested that customer and coworker psychological aggression are differently associated with personal outcomes, while coworker psychological aggression has no effects on fear.
Further studies show a positive relationship between workplace psychological aggression and irritation and depressive mood (Yang, Liu, Nauta, Caughlin, & Spector, 2016) and a negative association with overall health (Schat & Frone, 2011).
No significant associations were found between workplace psychological aggression and sadness, cynicism, needle stick injuries, allergies, tiredness, sleeplessness, and headache or the positive outcome of posttraumatic growth.
Workplace psychological aggression prevention strategies
Only 14 out of 30 studies examined means of preventing or buffering the consequences of workplace psychological aggression, and very few of them explicitly considered moderating variables (Table 3).
Correlations and moderators effects of individual and organizational variables related with positively strategies and resources to deal with psychological aggression.
p < .05. **p < .01. ***p < .001.
Of those 14 studies, just 5 examined moderating effects. One study (Zhou et al., 2015) focused on employees’ political skills (i.e. ability to influence others’ perceptions and behaviors). Political skills are a significant variable that could buffer the impact of workplace psychological aggression on anger, job satisfaction, career commitment, and injury. Three studies suggested that confidence to prevent and respond to workplace psychological aggression (Hanson et al., 2015), job resources (Viotti et al., 2015), and perceived organizational support (Li & Zhou, 2013) moderates the impact of customer psychological aggression on burnout. Support is also suggested by Schat and Kelloway (2003) as a significant moderating variable. Their results demonstrated that instrumental support moderates the effects of psychological aggression at work on emotional well-being, somatic health, and job-related affect. Informational support has moderating effects on emotional well-being.
Five studies identified a perceived violence-prevention climate (i.e. organizational policies, practices, and procedures to prevent physical and verbal aggression; Kessler et al., 2008) as a key variable with a significant correlation with the experience of direct and vicarious aggressive behaviors and strains. More specifically, a violence-prevention climate is significantly associated with psychological aggression, anger, anxiety, depression, physical symptoms, turnover intentions and emotional exhaustion (Chang et al., 2012; Kessler et al., 2008; Spector, Coulter, Stockwell, & Matz, 2007; Spector et al., 2015; Yang & Caughlin, 2017), management satisfaction, job satisfaction, and perceived danger (Kessler et al., 2008; Spector et al., 2007; Yang & Caughlin, 2017). In this context, aggression-preventive supervisor behavior and aggression-preventive employee effort have a significant correlation with psychological aggression and emotional exhaustion, anger, anxiety depression, turnover intentions, job management, and job satisfaction (Chang et al., 2012; Yang & Caughlin, 2017).
Two other studies also examined organizational support as an important resource to deal with psychological aggression. Support received from supervisors is negatively associated with customer psychological aggression, emotional exhaustion, and depersonalization and positively associated with personal accomplishment (Dormann & Zapf, 2004). A low-quality relationship between supervisor and subordinates (i.e., Leader–member exchange) and perceived organizational support are associated with psychological aggression (Chang & Lyons, 2012).
Chang and Lyons (2012) and Schat and Frone (2011) also suggested that organizations that implement strategies to promote job satisfaction and affective commitment will reduce psychological aggression and strains.
Autonomy is another organizational resource that is negatively associated with psychological aggression, emotional exhaustion, and stress appraisal resulting from customer aggression (Grandey et al., 2004).
Discussion
Workplace psychological aggression is a serious problem for employees and organizations and occurs frequently (ILO, 2013; Spector et al., 2014).
Based on current empirical data, we summarized the impact of workplace psychological aggression on employees’ health. This review goes further by providing significant findings related to resources and strategies to prevent and buffer the negative effects of psychological aggression at work. We examined workplace psychological aggression as a specific construct (cf. Nielsen & Einarsen, 2018), not psychological aggression at work as broad concept that encompass constructs as bullying or incivility. From the target’s perspective, we considered behaviors such as being yelled, shouted, or sworn at, which are perpetrated, directly or vicariously, by customers, supervisors, and coworkers toward employees.
Two decades have passed since the early investigations of Barling and colleagues (e.g., Barling, 1996), but critical limitations in the scientific research on workplace aggression persist, namely, the definition of the construct under investigation and the use and reporting of valid and reliable measures of exposure, making statistical comparisons and meta-analyses difficult. These methodological concerns are also found in other reviews (e.g., Spector et al., 2014).
It would be important, therefore, for future research on workplace aggression to define the form of aggression that is examined, identify the measure and the authors, clarify the behaviors nested, and, if the instrument has one item, report the item and specify other information such the source and time frame.
Workplace psychological aggression outcomes
According to the traditional model of work stress, workplace psychological aggression is a psychosocial stressor that affects employees through a psychological stress process, as opposed to a directly physical association (Barling, 1996; Ganster & Rosen, 2013; Schat & Kelloway, 2005). A widely used procedure to examine the consequences of workplace aggression is mediation analysis (e.g., Barling et al., 2001) which is extensively recommended in research (e.g., Hayes, 2017; Hayes & Rockwood, 2017; MacKinnon, 2011; Preacher, 2015).
Our systematic review only reported direct relationships between workplace psychological aggression and personal strains, not direct and indirect outcomes (mediating effects).
Our findings show that participants from a great variety of occupations, working in different settings, who have experienced psychological aggression also reported a multiplicity of negative outcomes.
Longitudinal associations show that employees experiencing psychological aggression have reported anger, depression, anxiety, and musculoskeletal injury during the study period (12 months). However, more prospective studies are required to examine how serious those outcomes became. In fact, Spector et al. (2015), when comparing strains over time for those exposed versus not exposed, found that those exposed to psychological aggression did not show an increase in strains.
Other results also demonstrated that overt psychological aggressive behaviors predict affective (fear) psychological (e.g., emotional exhaustion, depersonalization) and psychosomatic (physical symptoms) outcomes.
Our findings agree with Dudenhöffer and Dormann’s (2015) meta-analytic results concerning the consequences of customer-related social stressors. The authors found that the associations between verbal aggression by customers and emotional exhaustion and depersonalization are mainly invariant across service jobs.
When the source of psychological aggression is considered, we do not know the differences in the impact of psychological aggression perpetrated by different sources on personal outcomes over time, and only three studies compared differences related with the experience of psychological aggression from different sources and outcomes. As provided by Spector et al. (2014), various sources accounted for workplace psychological aggression (e.g., nurses, 21.8%; physicians, 28.5%; staff, 39.2%). Depending on the source, workplace psychological aggression is associated with different outcomes and differences in the impact on those outcomes (see LeBlanc & Kelloway, 2002). Therefore, a deeper examination of the target and perpetrator relationship will enable us to refine our strategies to minimize the impact of workplace psychological aggression on personal and organizational outcomes (Hershcovis & Barling, 2007).
None of the studies included addressed the consequences of vicarious psychological aggression. This is not surprising since traditionally research has examined the consequences of vicarious physical violence (Barling, 1996; Schat & Kelloway, 2005; see also Dupré et al., 2014). More research is needed taking this perspective into account to assist organizations in developing guidelines to prevent and minimize its effects (cf., for example, International Labor Office/International Council of Nurses/World Health Organization/Public Services International, 2002 or Occupational Safety and Health Administration, 2016 guidelines).
Of the studies included, few examined direct and indirect outcomes of workplace psychological aggression. Additionally, considering the reasons given by Hayes (2017), MacKinnon (2011), and Preacher (2015), we would encourage future research to include mediation variables in studies on workplace aggression outcomes.
Workplace psychological aggression: prevention and reduction of effects
The consequences of workplace psychological aggression are also influenced by individual responses and situational and individual factors that could prevent or minimize its effects (Barling, 1996; Schat & Kelloway, 2005). The second purpose of this review was to summarize empirically based evidence that can be relevant when designing preventive programs. Of the total of 30 studies, 14 studies provided the required information from different statistical approaches and supported by different theoretical models, for example, the conservation of resources theory (Li & Zhou, 2013).
There is little empirical research analyzing effective ways to moderate the relationship between workplace psychological aggression and outcomes. Nevertheless, the findings are promising, and relevant individual and organizational resources are advanced. More specifically, political skills, confidence to prevent workplace psychological aggression, job resources, and organizational support moderate the effect of workplace psychological aggression on outcomes. Individuals who develop interpersonal resources—interpersonal influence, networking ability, apparent sincerity—tend to experience lower levels of anger and injury after experiencing psychological aggression at work (Zhou et al., 2015). Strategies that increase employees’ confidence to prevent and respond to aggressive psychological behaviors may help workers to deal with those negative acts and decrease the effect of psychological aggression on burnout (Hanson et al., 2015). The development of strategies that increase resources at the job content level (work meaning, skill discretion, role clarity); at the social level (support from supervisors and colleagues); and at the organizational level (organizational support, fairness, social utility) have a significant impact on reducing emotional exhaustion and depersonalization (Viotti et al., 2015).
The role played by support from the organization, supervisor, and coworkers, at the individual and unit level, in preventing workplace psychological aggression and reducing its effects has been also demonstrated. Actions that indicate that organizations care about employees’ well-being and value their contributions (emotional, instrumental, and informational support) have a significant moderated effect on the negative consequences of psychological aggression (Li & Zhou, 2013; Schat & Kelloway, 2003).
Other strategies and resources include a violence-prevention climate, prevention behaviors, support, employee morale, and autonomy. Spector and colleagues focused on employees’ perceptions of a violence-prevention climate at the individual and at the unit level. Respectively, both policies (e.g., formal rules) and practices (e.g., management attitudes and support) encourage employees to focus attention on how their preventive behavior may influence the behavior of others (coworkers and patients). Organizations that promote a good violence-prevention climate enhance employees’ knowledge and help them to recognize precursors of violent and aggressive behaviors and take actions to avoid them. In addition, interventions aiming to promote a good perceived violence-prevention climate are associated with a decrease in individual (anger, anxiety) and organizational (turnover intentions) strains. Supervisors’ daily procedures and practices and employee prevention motivation and behaviors are also crucial for successful prevention and reduction of the effects of workplace psychological aggression. Aggression-preventive supervisor behavior (helping–serving, monitoring–controlling, instructing–guiding); motivation (e.g., being motivated to try to stop violent incidents at work); compliance (e.g., following the correct violence-prevention procedures); and participation (e.g., offers to help others by teaching them necessary knowledge or skills related to violence and aggression prevention) are significantly associated with a reduction of psychological aggression and outcomes (Chang et al., 2012; Yang & Caughlin, 2017).
Dormann and Zapf (2004) and Chang and Lyons (2012) also found evidence of the direct effect of social support on workplace psychological aggression and outcomes. The development of interventions related with different types of support provided by organizations and with promoting high-quality relationships between subordinates and their supervisors should be considered when designing programs to prevent and minimize the effects of workplace psychological aggression.
High levels of management satisfaction, job satisfaction, affective commitment, and autonomy are also suggested as positive organizational resources related with health promotion and prevention of workplace psychological aggression (Chang & Lyons, 2012; Grandey et al., 2004; Schat & Frone, 2011).
Limitations
Several limitations are considered. Our first limitation is due with critical appraisal of the evidence of the included studies that was not rated. With few exceptions (e.g., Nielsen, Indregard, & Øverland, 2016), this lack of information is usual in reviews (e.g., Tepper et al., 2017) or meta-analytical reviews (Spector et al., 2014) in the scientific research related with aggressive behaviors at work. The review team adapted a checklist (Checklist for Analytical Cross-Sectional Studies) from The Joanna Briggs Institute to ensure some consistency during the process of analyzing the studies (Moola et al., 2017). However, this tool, as well as assessing the quality of evidence, made it possible to make an overall evaluation of each study, based on certain criteria (e.g., existence or not of well-defined inclusion criteria, the existence or not of a description of the participants), ensuring their inclusion in this review.
Second, a lack of conceptual explanation and the heterogeneity of measuring instruments make it difficult to identify the content of psychological aggression measures, which may lead to bias regarding the content measured (see, for example, Bowling et al., 2015).
Third, although some studies tested direct and indirect predictive effects, those pathways were not analyzed due to the characteristics and length limitation of this systematic review, meaning that important patterns may not have been discussed. For example, in Schat and Kelloway’s (2003) study, the moderating effect of organizational support was stronger on secondary (e.g., emotional well-being) than on primary (fear) outcomes.
An additional limitation is that our findings are mainly from cross-sectional self-report studies, leaving uncertain the direction of causality in the predictions. This is a limitation commonly pointed out in research on aggression at work (see Nielsen et al., 2017; Tepper, 2007). However, research on the predictors of workplace aggression link, for example, negative affect (sadness, anxiety) and trait anger to aggressive behaviors at work (Barling et al., 2009; see also Barling, 1996; Douglas & Martinko, 2001).
Conclusion
A systematic review on workplace psychological aggression was conducted, and PRISMA guidelines (Moher et al., 2009) were followed to report the data in the manuscript. Focusing on empirical studies, our aim was to identify personal outcomes and ways to reduce and minimize its impact on employee well-being. The challenges faced when investigating workplace aggression are to clearly define our construct, use valid and reliable measures, and explore strains and moderators with valid and reliable instruments within a statistical analysis framework that provides us with better knowledge and understanding of this phenomenon. These issues should be interpreted with some reservations, bearing in mind, for example, the scientific discussion in this field related with formative and reflective measures (e.g., Dionisi et al., 2012; MacKenzie, Podsakoff, & Podsakoff, 2011; Tepper & Henle, 2011) or single-item measures (e.g., Gilbert & Kelloway, 2014). Regarding the severe outcomes of workplace psychological aggression, negative associations were found with personal outcomes across samples.
Little is known about how and when the variability of its effects occurs at the supervisory, coworker, and customer level, and more longitudinal associations are needed to conclude about causality.
Programs to prevent workplace psychological aggression should develop strategies related, for example, with aggression-preventive supervisor and employee behavior, political skills, job resources, or organizational support. This is in line with the APA’s (2014) recommendations “that preventive programs be selected based on a careful review of empirical evidence in order to choose programs that are empirically supported for their specific contexts and specified goals” (p. 287).
Practical implications
The findings of this systematic review emphasized workplace psychological aggression as a severe problem for individuals showing that those who experience psychological aggression at work also experience significant negative physical and psychological outcomes.
Importantly, this review also has implications for organizations by providing evidence that there are ways to moderate the impact of psychological aggression on personal outcomes and organizational strains. For example, (i) enhance organizational support through programs to assist those that experience psychological aggression, improve leadership skills, and provide stress management training, (ii) the development of programs aimed to create a positive violence-prevention climate through training, having practices that respond quickly to aggression and provide relevant information on workplace psychological aggression, or (iii) develop political skills through mentoring relationships (e.g., Chang & Lyons, 2012; Chang et al., 2012; Schat & Frone, 2011; Zhou et al., 2015).
In doing so, organizations will promote individuals’ well-being, life satisfaction, and career commitment. As a result, workers will be more able to interact with other members of the public, relatives, coworkers, supervisors and influence their behaviors positively (Spector et al., 2007). Based on existent evidence, this systematic review indicated several directions for future interventions, programs, and guidelines.
If our findings are relevant for occupational services, preventive interventions, and health promotion at work, they are also relevant for the general area of occupational psychology and professional psychology practices (psychotherapy, assessments, and psychological interventions). Following APA (2016) recommendations for practice, clinical interventions related to work and career should be implemented to promote employees’ health and quality of life. In fact, work plays a critical role in terms of time engagement and psychological meaning, is associated with satisfaction in other domains of life, and may have a protective function for adverse life events.
Footnotes
Acknowledgements
We thank Andrew Thomson for his English review.
