Author	Affiliation
Sarayna S. McGuire, MD	Mayo Clinic, Department of Emergency Medicine, Rochester, Minnesota
Aidan F. Mullan, MA	Mayo Clinic, Department of Quantitative Health Sciences, Rochester, Minnesota
Casey M. Clements, MD, PhD	Mayo Clinic, Department of Emergency Medicine, Rochester, Minnesota

Introduction
Methods
Results
Discussion
Limitations

INTRODUCTION

Background

Workplace violence in healthcare is a serious threat to staff. Between 2011–2013, the number of workplace assaults averaged approximately 24,000 annually, with nearly 75% occurring in healthcare settings. Data from the Bureau of Labor Statistics show that incidents of serious workplace violence were four times more common among healthcare workers than those in private industry.¹ Emergency departments (ED) and psychiatric hospitals are two areas in healthcare where violence is most commonly reported.

Importance

There is a concerning rise in ED violence, with one in five ED directors reporting guns or knives brought into the ED on a daily or weekly basis.² Violence against healthcare workers continues to make the news, and hospital-based shootings nearly doubled between 2000–2011.³ Among ED nurses, prior research has shown an annual incidence of verbal and physical abuse ranging from 39–98% and 13–67%, respectively.^4,5 Among emergency physicians, the incidence has ranged from 75%–96% and 51%–78%, respectively.^1,4,5 However, this likely represents under-reported data as only 30% of nurses and 26% of physicians go on to report incidents of violence.⁶ According to a 2018 poll from the American College of Emergency Physicians, nearly 70% of respondents believed that violence in the ED has increased during the previous five years and nearly 80% felt that patient care was affected as a result.⁷

Goals of This Intervention

Exposure to workplace violence impacts the entire team; however, there is a paucity of research evaluating the incidence of violence experienced by the ED multidisciplinary care team and how it compares to institutional reporting. We sought to survey all staff at our academic ED to identify the incidence of verbal abuse and physical assault over a six-month period and compare responses to documented incident reports from the same time period to evaluate for under-reporting of violence. We also sought to obtain baseline characteristics of respondents to evaluate for risk factors for violence or under-reporting. We hypothesized that nearly all members of the ED multidisciplinary care team have been exposed to verbal abuse over a six-month time period, with many of these incidents going under-reported, and that a significant percentage of staff have also experienced physical assault during the same time frame.

METHODS

Study Design and Setting

This descriptive prospective study took place between April–May 2020 within the ED of a large, academic, Level 1 trauma center in a small urban city in the Midwest. The ED sees an average of 78,000 patients annually and has 24/7 security presence available.

Selection of Participants

The target population consisted of all ED staff, including clinicians (attending and resident physicians, and advanced practice providers [APP]), nursing, care team assistants (CTA) who provide clerical support and limited patient interaction, patient care assistants (PCA), electrocardiogram (ECG) and radiology technicians, phlebotomists, registration staff, and security officers who worked in the ED at least six months prior to taking the survey. After review by the institutional review board (IRB), the survey (described below) was emailed to all distribution lists for the abovementioned target population with a cover letter describing the study purpose, directions for participation, and information regarding informed consent. The questionnaire included a statement of informed consent at the beginning, and completion indicated participant consent for inclusion in the study. Two reminder notices were sent two weeks apart through the same method. The IRB reviewed this study and materials and deemed it exempt from approval requirement.

Measurements

We developed an anonymous REDCap survey (Research Electronic Data Capture, Vanderbilt University, Nashville, TN)⁸ that included both multiple-choice and Likert-scale response questions. This survey was distributed broadly by department and job type to anyone who might work in the ED, even occasionally. Respondents were asked to self-select for if they had done any work in an ED in the preceding six months. Participants were asked to indicate whether they had experienced any of the following forms of verbal abuse in the prior six months (October/November 2019–April/May 2020) while working in the ED: threatening tone of voice; abusive language/statement; harassment (eg, racial, gender, sexual); or personal verbal threats (eg, threat of physical or sexual violence, threat of physical assault to occur outside the workplace). Participants were asked to indicate whether they had experienced any of the following forms of physical assault in the prior six months while working in the ED: physical assault with weapons (including hospital equipment); physical assault with bodily fluids (eg, saliva, urine, feces, wound exudate, blood, or spit); or physical assault in the form of punching, biting, rough handling, scratching, kicking, shoving/pushing, or hitting. If answering affirmatively to any of these choices, respondents were asked to indicate whether they had formally reported the incident.

 

Population Health Research Capsule

What do we already know about this issue?

Serious workplace violence incidents are five times more common in healthcare than other industries, and events are often not reported.

What was the research question?

What is the incidence of workplace violence in the emergency department, risk factors, and why is it under-reported?

What was the major finding of the study?

Most staff, including support disciplines, experienced violence and most incidents went unreported.

How does this improve population health?

As health systems seek to improve the safety of their employees, they must direct initiatives to the entire healthcare team since no group is immune.

 

We used Likert scales to measure participants’ perceptions of safety and estimated frequency of verbal abuse, physical assault, and reporting of incidents of workplace violence in the prior six months. Standard demographic measures were collected, including gender, profession, primary shift worked, and years of experience, and whether the employee had worked in the ED for at least six months. We asked the institution’s Office of Security to provide de-identified data on the number of verbal abuse and physical assault incident reports filed by ED staff during the same time period for comparison.

Outcomes

The primary outcome was the incidence of verbal abuse and physical assault experienced by ED staff in a six-month time period as indicated by survey responses. The secondary outcome was the comparison of this self-revealed data to formally reported incidents during the same time period.

Data Analysis

Survey responses were summarized with frequency counts and percentages. We performed group comparisons of survey responses using chi-squared tests. Pairwise group comparisons were performed using odds ratios calculated from frequency counts. Confidence intervals were generated using asymptotic Gaussian approximation. We converted Likert-scale responses to the perceptions of safety question to a numeric rank based on the strength of sentiment. A two-sided Wilcoxon rank-sum test and Kruskal-Wallis test were performed on these ranks to compare responses between gender and years-of-experience groups, respectively. All tests were two-sided and P-values less than 0.05 were considered significant. We performed analyses using R version 3.6.2 (The R Foundation for Statistical Computing, Vienna, Austria).

RESULTS

Characteristics of Study Subjects

A total of 261 responses were received. Seventeen respondents indicated working in the ED less than six months and two respondents indicated working in management with no clinical duties—these 19 responses were excluded from analysis. We included the 242 remaining responses in our analysis (Table 1). The cohort was 59.5% female. The most common positions were nursing staff (80/242, 33.1%), security (40/242, 16.5%), and attending physicians (28/242, 11.6%).

Violence by Position

Overall, 208 (86%) respondents indicated they had been verbally abused in the preceding six months (Table 2). Security officers had the highest incidence of verbal abuse (98%), followed by nursing (95%), PCAs (90%) and clinicians (90%), phlebotomists (75%), CTAs (73%), registration staff (50%), and ECG/radiology technicians (50%). Non-security and non-nursing personnel indicated an incidence of verbal abuse of 78%, which was significantly lower than either security (odds ratio [OR] = 0.08, 95% confidence interval [CI], 0.01 – 0.62, P <.001) or nursing staff (OR = 0.17, 95% CI, 0.06 – 0.50, P < .001).

Staff indicated how often they were verbally abused by patients or family members in the prior six months (Table 3). Security personnel had the highest proportion of responses indicating incidents of verbal abuse at least every week (16/40, 40%), followed by nurses (30/80, 38%). For non-security and non-nurse employees, only 11% of respondents indicated verbal abuse occurring at least every week, which was significantly lower than either security (OR = 0.19, 95% CI, 0.08 – 0.45,P < .001) or nursing staff (OR = 0.22, 95% CI, 0.11 – 0.44, P < .001).

Overall, 90 (37.2%) respondents indicated that they had been physically assaulted in the preceding six months (Table 4). Security officers had the highest incidence of physical assault (73%), followed by nursing (49%), PCAs (30%), clinicians (24%), phlebotomists (17%), and ECG/radiology technicians (13%). Neither CTAs nor registration staff revealed any physical assault. Again, security had the highest frequency of assault, with 29 of 40 (73%) respondents indicating being physically assaulted at least once. Nurses had the next highest frequency of assault (39/80, 49%). For non-security and non-nurse staff, 22 (18%) respondents indicated at least one incident of physical assault. This frequency was significantly lower than security (OR = 0.08, 95% CI, 0.04 – 0.19, P < .001) and nursing staff (OR = 0.23, 95% CI, 0.12 – 0.44, P < .001).

Table 5 describes the frequency of reporting events of workplace violence, grouped by position. Security personnel had the lowest proportion of respondents indicating they never report incidents, with seven (18%) responding in this way. Comparatively, 140 (69%) non-security personnel responded that they never report incidents. The odds that a non-security staff member responded “Never” were 11 times higher than for security personnel (OR = 10.65, 95% CI, 4.47 – 25.38, P < .001).

Violence by Gender

Table 6 provides the number of respondents experiencing verbal abuse, grouped by gender. Overall, there was no difference in the incidence of verbal abuse between genders (female: 85%; male: 87%, P = 0.70). Males were more likely to report incidents of verbal abuse compared to females (OR = 3.87, 95% CI, 1.77 – 8.47, P < .001). However, once we account for employee position, there was no difference in reporting between males and females. For security personnel, 16/29 (55%) males and 4/6 (67%) females indicated reporting verbal abuse experienced (OR = 1.63, P = 0.61). For non-security personnel, 7/53 (13%) males and 7/116 (6%) females indicated reporting the abuse (OR = 2.37, P = 0.12).

Table 7 summarizes the incidence of physical assault. There was no significant difference in the overall incidence of physical assault between genders (female: 33%; male: 43%, P = 0.16). However, males experienced 2.8 times more occurrences of assault with bodily fluids compared to females (OR = 2.82, 95% CI, 1.43 – 5.55, P = .002). Males who experienced physical assault were more likely to report the incident compared to females (OR = 3.79, 95% CI, 1.57 – 9.18, P = .003). Again, there was no difference in reporting between males and females after accounting for employee position. Among security personnel, 19/21 (90%) males and 5/6 (83%) females indicated reporting physical assault experienced (OR = 1.9, P = 0.63). For non-security personnel, only 6/20 (30%) males and 9/41 (22%) females indicated reporting physical assault experienced (OR = 1.52, P = 0.5).

Violence by Shift

There was no difference in the overall incidence of verbal abuse between shifts (x² = 4.63, P = .20); However, staff working during the evening reported 69% fewer instances of abusive tone (OR = 0.31, 95% CI, 0.15 – 0.64, P = .001). Staff working day or overnight shifts were 4.2 times more likely to report incidents of verbal abuse compared to those working evening or rotating shifts (OR = 4.17, 95% CI, 1.85 – 9.39, P < .001). There was no significant difference in physical assault related to shifts (x² = 3.97, P = .26). Moreover, there was no significant difference in the number of respondents reporting incidents of assault (x² = 7.01, P = .071).

Violence by Years of Experience

Staff members with less than four years or more than 21 years of experience were more likely to experience some form of verbal abuse compared to staff members with 5–20 years of experience (OR = 2.94, 95% CI, 1.31 – 6.61, P = .007). There was no difference in the number of respondents reporting their incidents of verbal abuse between years of experience (x² = 4.18, P = .24). There was no difference in the number of respondents indicating some form of physical assault between experience groups (x² = 6.00, P = 0.11). Additionally, there was no difference in the number of respondents reporting physical assault between experience groups (x² = 2.02, P = .57).

Perceptions of Safety

When asked how safe respondents felt as a staff member working in the ED, 100% of respondents indicated subjectively feeling safe, with 11.1% indicating feeling extremely safe (27/242), 48.8% very safe (118/242), 35.5% moderately safe (86/242), and 4.5% slightly safe (11/242). No respondents indicated feeling unsafe. Responses were converted to a numeric rank based on the strength of sentiment, with “slightly safe” the lowest score at 1 and “extremely safe” the highest score at 4. Males had a higher perceived safety compared to females (P = .016). The average response rank for males was 2.81, compared to 2.56 for females. When evaluating for perceptions of safety among staff with different years of experience, there was no significant difference in perceived safety between the experience groups (P = .40).

Official Incident Reports

During the same six-month time frame respondents were surveyed, there were only 11 official incident reports made to the Office of Security regarding verbal threats or harassment and 18 reports of physical assaults. Compared to self-reported data from the survey, this corresponds to a 5% and 18% reporting rate, respectively.

DISCUSSION

Similar to prior research, our survey of ED staff showed a high incidence of verbal abuse (86%) and physical assault (37%) within our ED over a six-month time period. Through surveying the entire multidisciplinary team, we were able to demonstrate that all team roles experienced verbal abuse at some point in a six-month time period, and nearly all experienced physical assault with the exception of CTAs and registration staff. Even so, interestingly all 242 respondents indicated feeling some degree of safety in our ED. We recognize that this subjective reporting of safety may be misleading and may be attributed to a selection bias as healthcare employees who feel unsafe in their workplace are more likely to transfer out of the department and may have been missed by our survey. This finding may also mirror prior literature that healthcare employees are resistant to the belief that they are at risk for patient-initiated violence and experience a complacency in thinking that violence is simply “part of the job.”¹

Security personnel were more likely to formally report incidents compared to non-security personnel victims. This may be due in part to the nature of their job and the frequency with which they experience violence, as well as familiarity with the reporting process as departmental incident reports are submitted to their office. Concerningly, 69% of non-security personnel indicated that they never report incidents of violence. This was corroborated with a review of official incident reports received during the same time period. Barriers to reporting are multifactorial and include, as described above, the belief that violence is “part of the job,” confusion over what constitutes violence, unfamiliarity with reporting processes, lack of available time at work for reporting incidents, fear of retribution from supervisors, and perceived lack of institutional support.^1,9 Our study findings indicate that future efforts to increase incident reporting within the ED should focus on the entire multidisciplinary team, including visiting staff assigned to non-ED departments such as phlebotomy, cardiac monitoring (ECG), and radiology.

In terms of isolating specific risk factors, we found no difference in the overall incidence of violence between genders; however, males were significantly more likely to report incidents of both verbal and physical abuse compared to females. To our knowledge, this has not been previously described in the literature. Although we found a difference in the reporting of violence between genders, this difference was not significant once we accounted for employee position. This is likely due to a greater proportion of males in our study working in security (35% males, 5% females), and security personnel indicating a higher rate of violence reporting regardless of gender. Future studies with larger cohort sizes should seek to identify whether there is a difference in reporting between genders. Coincidentally, females had a significantly lower perceived perception of safety in our ED compared to their male counterparts. Thus, additional research should seek to more clearly establish the reasons why more females choose not to report incidents of violence.

There was no significant difference in the overall incidence of violence between shifts; however, staff working daytime or overnight shifts were more likely to report incidents of verbal abuse. This may be explained by the higher frequency of incident reporting by security staff and the fact that security officers in our institution work 12-hour shifts, considered either day shift or overnight shifts, with only non-security personnel working evening or rotating shifts. Future research should continue to distinguish what additional demographic factors may be contributing to the lack of violence reporting.

The unique environment of the ED contributes to its propensity for violence: stress among patients, families, and visitors; long wait times and delays; crowding; unrestricted 24-hour access; low socioeconomic status; substance abuse; patients with behavioral health issues; gang activity; and frequent delivery of “bad news” have all been suggested to contribute to the elevated incidence of violence.^1,10 A multidisciplinary study of healthcare workers found exposure to workplace violence significantly correlated with burnout, and a separate survey of ED nurses found that 94% of those experiencing violence in the workplace exhibited symptoms of post-traumatic stress.¹¹ In addition to its impact on patient care and detriment to employee wellbeing, violence has a substantial financial impact for employers and the economy. Financial costs of workplace violence include lost time/wages; medical costs of employee injury, disability, and/or death; and attrition.¹² According to a 2017 report commissioned by the American Hospital Association, hospitals spent an estimated $1.1 billion in security and training costs to prevent violence, and an additional $429 million to cover costs such as medical care, staffing, and insurance resulting from violence against staff.^13,14 Future research should attempt to characterize the mental and physical toll on the multidisciplinary ED care team to help direct efforts for employee wellbeing.

This study’s findings have important clinical implications. The incidence of verbal abuse among our multidisciplinary ED care team was nearly 6 of every 7 staff members, and yet these incidents were almost never reported to the institution. The incidence of physical assault was more than 2 of every 5 staff members and, again, the majority went unreported. Nearly 7 out of every 10 non-security staff members declined to officially report the violence they experienced. Findings from this study suggest that the pervasive nature of violence in healthcare is still underappreciated and that increased efforts are needed to protect ED staff members and support and encourage or incentivize accurate and reliable reporting.

LIMITATIONS

This study has several important limitations. To preserve anonymity of employees, the study was sent to email distribution lists (DL) and included some DLs with employees working in other departments other than the ED (eg, phlebotomy, and ECG and radiology technicians), or who also worked at additional sites elsewhere in our health system (eg, clinicians). Thus, it is not possible to know the actual number of employees from different disciplines who work in the ED to estimate a response rate for our survey. Additionally, to further preserve anonymity, we did not ask in-depth demographic questions. Without knowing full-time vs part-time status of respondents, it is possible that some responses came from employees working part time and this may have skewed our incidence of violence. The definition of “verbal abuse” is highly subjective to individual respondents and survey inclusion of “threatening tone of voice’” may have contributed to over-reporting of verbal abuse in general by respondents.

The study was also subject to recall and reporting bias in terms of violence experienced over a six-month time period, as well as the reporting of incidents. We acknowledge that because this was a single-center study some aspects may not be generalizable to all institutions or geographic regions. However, the finding of under-reporting is not dissimilar to other published studies,^15,16 and the fact that abuse and violence affect previously unstudied populations including ancillary services and clerical assistant staff is important and not likely related to local factors.

Footnotes

Section Editor: Gary Johnson, MD

Full text available through open access at http://escholarship.org/uc/uciem_westjem

Address for Correspondence: Sarayna S. McGuire, MD, Mayo Clinic, Department of Emergency Medicine, 200 First St SW, Rochester, MN 55905. Email: McGuire.Sarayna@mayo.edu. 5 / 2021; 22:702 – 709

Submission history: Revision received September 26, 2020; Submitted January 30, 2021; Accepted February 17, 2021

Conflicts of Interest: By the WestJEM article submission agreement, all authors are required to disclose all affiliations, funding sources and financial or management relationships that could be perceived as potential sources of bias. No author has professional or financial relationships with any companies that are relevant to this study. There are no conflicts of interest or sources of funding to declare.

REFERENCES

1. Philips JP. Workplace violence against health care workers in the United States. N Engl J Med. 2016;374:1661-9.

2. Kansagra SM, Rao SR, Sullivan AF, et al. A survey of workplace violence across 65 U.S. emergency departments. SAEM. 2008;15:1268-74.

3. Weyand JS, Junck E, Kang CS, et al. Security, violent events, and anticipated surge capabilities of emergency departments in Washington state. West J Emerg Med. 2017;18(3):466-73.

4. Gerberich SG, Church TR, McGovern PM, et al. An epidemiological study of the magnitude and consequences of work related violence: the Minnesota Nurses’ Study. Occup Environ Med. 2004;61:495-503.

5. Gates DM, Ross CS, McQueen L. Violence against emergency department workers. J Emerg Med. 2006;31:331-7.

6. Behnam M, Tillotson RD, Davis SM, et al. Violence in the emergency department: a national survey of emergency medicine residents and attending physicians. J Emerg Med. 2011;40(5):565-79.

7. Kelly M. Violent attacks against emergency physicians remain a problem. Ann Emerg Med. 2020;75:11A-4A.

8. Harris Paul A, Taylor Robert, Thielke Robert, et al. Research electronic data capture (REDCap) – A metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377-81.

9. Stene J, Larson E, Levy M, Dohlman M. Workplace violence in the emergency department: giving staff the tools and support to report. Perm J. 2015;19(2):e113-7.

10. Lenaghan PA, Cirrincione NM, Henrich S. Preventing emergency department violence through design. J Emerg Nurs. 2018;44:7-12.

11. Copeland D, Henry M. The relationship between workplace violence, perceptions of safety, and professional quality of life among emergency department staff members in a Level 1 trauma centre. Int Emerg Nurs. 2018;39:26-32.

12. Larson LA, Finley JL, Gross TL, et al. Using a potentially aggressive/violent patient huddle to improve health care safety. Jt Comm J Qual Saf. 2019;45:74-80.

13. Taylor JL, Rew L. A systematic review of the literature: workplace violence in the emergency department. J Clin Nurs. 2010;20:1072-85.

14. Durkin M. Hospital fight back against violence. 2017. Available at: https://acphospitalist.org/archives/2017/12/hospitals-fight-back-against-violence.htm. Accessed May 25, 2020.

15. Pompeii LA, Schoenfisch AL, Lipscomb HJ, et al. Physical assault, physical threat, and verbal abuse perpetrated against hospital workers by patients or visitors in six U.S. hospitals. Am J Ind Med. 2015;58(11):1194-204.

16. Speroni KG, Fitch T, Dawson E, et al. Incidence and cost of nurse workplace violence perpetrated by hospital patients or patient visitors. J Emerg Nurs. 2014;40(3):218-28.