Violence against health care workers is not a novel phenomenon. The Occupational Safety and Health Administration (OSHA) defines workplace violence as “violent acts, including physical assaults and threats of assault, directed toward persons at work or on duty.”1
Health care workers are more susceptible to violence in the workplace than any other group of workers in the United States. From 2002 to 2013, the Bureau of Labor Statistics data indicate that workers in health care-related fields suffer workplace violence injuries that require taking time off work at a rate 4 times that of other industry workers.2 According to a 2016 OSHA report, ∼24,000 workplace assaults occurred annually in health care settings between 2010 and 2013, resulting in major and minor physical injury, psychological harm, temporary or permanent physical disability, and death.2 Health care workers make up 13% of the total workforce, but ∼60% of all workplace violence incidents take place in hospitals, clinics, emergency rooms, and other health care facilities.3,4 This violence against health care workers includes not only verbal and physical abuse but also threats against their lives.
One might assume that anesthesiologists practice in very controlled, well-synchronized environments, immune to the threat of violence from patients or colleagues. However, the perioperative environment, intensive care units (ICUs), and pain clinics are high-emotion, high patient/family member stress environments, a known risk factor for violence against workers.
Scope of the Problem
Accurate quantification of violence against health care workers is difficult for a variety of reasons. First, official statistics only capture injuries that require at least 1 day away from the workplace, making nonverbal and less aggressive threats difficult to measure.1 Second, there is reluctance among health care workers to report workplace violence to a manager or an administrator. A survey of 7 medical centers showed that 62% of respondents experienced violence in the previous 12 months, but fewer than 25% reported those events. When they do report, workers are much less likely to report verbal assaults compared to physical assaults. Reasons include a workplace culture that dissuades reporting, fear of making the organization look bad, fear of retaliation or demotion, and the mindset that these assaults are “part of the job.” Furthermore, victims surveyed report that they do not have faith that their concerns will be taken seriously and are rarely satisfied with the outcome of reporting.1,4,5
Worker safety appears to assume a lower priority compared with patient safety in hospitals. Resourcing is scarce, with very few comprehensive workplace violence-prevention programs in health care organizations. Even when present, programs are sometimes designed and messaged with patient safety—not worker safety—as the primary goal. Few track the number of continuous violence-free workplace days in an analogous manner to the “merit boards” that track continuous hospital-acquired infection-free patient days. Only 23% of hospital boards include worker safety in their board of directors’ dashboards.6 Currently, there are no federal or national mandates [eg, OSHA, The Joint Commission (TJC)] and few statewide or regional regulatory bodies that require health care organizations to report on workplace violence. Therefore, current workplace violence data remain anecdotal and survey based.5
Within hospitals, workplace violence in emergency department and psychiatric units is the most commonly reported and studied. For emergency departments, reported risk factors include protracted wait times, crowded waiting rooms, poor-quality food options, being given bad diagnostic or prognostic news, and the presence of weapons.7 Many of these same risk factors can occur in operating rooms, ICUs, and pain clinics. For example, surgical patients and their loved ones can be under a significant amount of emotional, financial, and physical stress that accompanies a new diagnosis of disease, recurrence or spread of cancer, or a decline in cognitive or physical function. Anesthesiologists in chronic pain management also work in emotionally charged environments that extend into the outpatient clinic setting. These factors place anesthesiologists, interventional pain specialists, intensivists, nurse anesthetists, and the entire ICU and operating room team at risk for verbal and/or physical assault.
Impact on Patient Care, Clinicians, and Hospital Systems
Clinician behavior and decision-making have become scrutinized more intensely by both institutions and patients. Clinicians want to help and heal patients, but a retaliatory response may occur to unintentional, and sometimes unavoidable, failures by the health care team. When clinicians fail to meet the expectations of the patients they are doing their best to help, there is the potential for demands, threats, physical, and verbal abuse.8 Workplace violence has the potential for a negative ripple effect that can adversely and irreversibly impact the medical professional, their significant others, coworkers, patients, and employer.
Impact on Patient Care
Occupational violence has been associated with increased turnover and absenteeism, reduced productivity, decreased staff morale, increased counseling utilization and costs, and poorer patient care outcomes.9 When physicians are verbally or physically assaulted by patients, a protective coping response may include becoming defensive and a tendency to withdraw from patient care. To reassert their own needs and protect themselves from further injury, the physician may not be as involved in the patient’s care and may not feel comfortable speaking freely to the patient and their family members.8 These patients, in turn, are less satisfied with their care and caregiver.1 In the end, this cycle erodes the mutual respect and trust necessary for optimal patient care.
Impact on Physicians
A multidisciplinary study carried out by Gascon et al10 found that health care workers exposed to workplace violence had a significantly higher rate of burnout. Burnout was characterized as emotional exhaustion, work inefficiency, and depersonalization. The stress of violence has been shown to significantly affect the cognitive workload demands of emergency department staff members, specifically their ability to concentrate and manage their workload.11 In addition, posttraumatic stress symptoms experienced by victims have been correlated to a decline in the provider’s communication skills, capacity for empathy, and emotional support of others.12 When a health care worker has suffered from abuse at the hands of a patient, the evidence supports that the worker is no longer capable of performing to their full potential, opening the door for patient care errors.
Impact on Health Care Systems
Not only are victims of workplace violence more prone to commit errors and be less productive, they may seek employment elsewhere to lower their risk of exposure to violence.13 This decreased retention and increased turnover has costs to the health care system (eg, recruitment, onboarding/training). The turnover can also result in lack of continuity of care experienced by patients.1,5
One cannot separate patient safety from the safety of health care providers. The profession of anesthesiology prides itself on vigilance as a guiding principle to ensuring the safety of our patients. We must maintain the same level of vigilance when acts of violence occur in the workplace. We must highlight the pervasiveness of the problem and structure viable solutions that protect both providers and patients equally.
Our review of the literature found an absence of data on the prevalence of workplace violence among anesthesiologists. Therefore, we conducted a survey to better understand the prevalence and incidence of workplace violence amongst anesthesiologists in their various work settings.
We developed a 22-question survey using Survey Monkey. The survey was approved by the American Society of Anesthesiologists (ASA) for distribution by the professional membership email. Active ASA members that do not opt out of surveys comprise a membership of 40,428—both US and international—including anesthesiologists in training and practicing anesthesiologists (MDs/DOs).
The survey was open from September 29, 2018 to November 7, 2018, with reminders sent on October 7 and 14, 2018.
Survey Statistical Methods
We created a multivariable logistic regression model for each of our 2 binary outcomes: physical violence and nonphysical violence. The model covariates included sex, age cohorts, country of practice, race, ethnicity, practice setting, and whether the anesthesiologist had also experienced the other form of violence (physical for the nonphysical model and nonphysical for the physical model). Statistical significance was defined as a P-value <0.05. Adjusted odds ratios (AORs) and 95% confidence intervals (CI) were calculated. Analysis was carried out using SAS software (Version 9.4; SAS Institute, Cary, NC).
Our survey yielded a total of 2694 responses. The average time to complete the survey was <2 minutes. Table 1 presents the general respondent demographics.
This survey indicated the prevalence of physical violence to be 20% (Table 2) among those responding, with a majority of incidents occurring once or >12 months ago (Fig. 1). The prevalence of nonphysical violence was more common at 69%. A significant percent of respondents (>20%) indicated that they experienced nonphysical violence at least quarterly (Fig. 1). The most common source of physical violence against the anesthesiologists who responded to our survey was patients/family members. Interestingly, the primary source for nonphysical violence was physicians in other specialties, followed by patients/family members (Fig. 2).
As with other surveys, a majority of our respondents did not report the event to a supervisor, human resources, hospital leadership, or law enforcement. Further, most did not take time away from work as a result of the incident and did not feel that the situation was addressed and resolved to their satisfaction (Table 2).
Seventy-five percent of the survey respondents answered that they had not received any training on deescalation of a threat in the workplace. Sixty-one percent responded that they had not received any training on what to do in an “Active Shooter/Silver Alert” threat in the workplace.
In analyzing the survey responses for which there were no missing data (no gaps in answering questions and no “prefer not to respond” responses), female anesthesiologists had a higher risk of reporting both nonphysical violence and physical violence. Anesthesiologists who practiced in a pain management practice were at a higher risk for experiencing nonphysical violence. It is noteworthy that anesthesiologists below 45 years of age had lower odds for reporting nonphysical violence compared with those physicians in the 46 to 55 age cohort. Anesthesiologists whose primary practice is non-US had a higher risk of reporting physical violence. Those who indicated their race as nonwhite were at a reduced risk for experiencing physical workplace violence. Finally, those who reported nonphysical violence were more likely to also report experiencing physical workplace violence; similarly, those reporting physical violence were more likely to report nonphysical violence (Table 3).
Although we had close to 3000 survey respondents, this number is still small compared with the overall number of anesthesiologists within the ASA. The respondents’ demographics may not mirror the demographics of anesthesiologists across the United States. As a voluntary survey, there may have been biases from those who chose to respond. As we have described above, under-reporting is likely. However, this is still the largest anonymous survey to date of anesthesiologists experiencing workplace violence. In addition, the survey prevalence rates mirror that of other specialty physicians (ie, emergency medicine and psychiatry).
Anesthesiologists report a relatively high incidence of workplace violence in this pilot survey. We hope that these findings will trigger expanded research and focus in this area. We hope that a study that includes anesthesiologists with other members of the perioperative team (eg Certified Registered Nurse Anesthetists, Certified Anesthesiologist Assistants, OR nurses, surgeons, technicians) will be carried out. There is an important opportunity for individuals, teams, and organizations to address this harm—a harm that impacts the safety of the individuals involved and patient safety.
Resources and Successful Strategies to Address Workplace Violence
Anesthesiologists are not immune to the important problem of workplace violence. What are the possible solutions? This complex issue requires both public and private leadership at all levels: organizational, team, and individual.
A discussion on organizational leadership and workplace safety necessarily should include the lessons from Paul O’Neill and his leadership at Alcoa. His 13-year tenure at Alcoa is legendary, with lost work days to injury per 100 workers dropping over 15-fold, while increasing the annual net income 5-fold. His 3 tenets are key to all organizational leaders, including those in health care.
Can everyone in your organization say YES to these 3 things:14
- Every day, I am treated with dignity and respect by everyone I encounter without respect to my pay grade, or my title, or my race, or ethnicity or religious beliefs or sex.
- I am given the things I need—education, training, tools, encouragement—so I can contribute to this organization that gives meaning to my life.
- I am recognized for what I do by someone I care about.
These 3 statements reflect worker safety and engagement. The first is particularly germane to workplace violence.
National, Regional, and State Resources
There are resources at the national level both in the United States and in Canada (Table 4). A bill was introduced last year in the US House of Representatives, H.R. 5223, The Health Care Workplace Violence Prevention Act. The Bill “requires the Department of Labor to address workplace violence in healthcare facilities pursuant to the Occupational Safety and Health Act of 1970. Specifically, Labor must issue a rule that requires certain healthcare employers to adopt a comprehensive plan for protecting health care workers and other personnel from workplace violence.”15 Unfortunately, the bill was not enacted.
As state laws have varying requirements,7 the reader should check relevant state laws and connect with their state hospital associations to work in partnership toward policymaking that can improve the recognition and reduction of workplace violence. Several state hospital associations have taken on a leadership role to advance this work. Minnesota, Massachusetts, and Oregon are such examples with tools readily available to all (Table 4).
Hospital Strategic Initiatives
Learning from other hospitals can help build a comprehensive workplace violence prevention program within one’s own hospital. Beth-Israel Deaconess Medical Center (BIDMC) has made workplace violence prevention and recovery a strategic imperative since 2014-2015. Their comprehensive program is aptly named Don’t Worry Alone. In 2014-2015, BIDMC established a Workplace Violence Prevention Committee consisting of broad representation from over 11 departments including patient safety, healthcare quality, nursing, hospital medicine, social work, public safety, employee/occupational health, interpreter services, communications, corporate compliance, and emergency management.
Over 3 years, BIDMC has engaged in a 4-pronged effort to recognize and mitigate workplace violence: a comprehensive security evaluation, improving the recognition and reporting of incidents, creating safer alerts and response teams, and standardized education. (Pat Folcarelli, RN, VP, Health Care Quality, personal oral communication, October 2018).
First, BIDMC engaged an outside consultant to perform a comprehensive security evaluation. This included extensive review of security plans, policies, and practices. In fiscal year 2018, a $1 million capital improvement program that included physical security and tighter access control was implemented.
Second, the hospital realized that workplace violence was likely under-reported. Their work showed that the primary reasons not to report were (1) “It’s part of the job,” (2) uncertainty in terms of what constitutes violence, (3) the patient’s clinical condition (rationalizing violence to be expected), (4) management accountability toward reporting (did not expect a response), (5) did not want to get involved, and (6) did not have time. BIDMC’s efforts to improve reporting are directed toward both ease and convenience of reporting. They widened the ability to report through multiple avenues. Nursing supervisor shift reports are a dominant source of incident reporting. As reporting increased, nonphysical workplace violence constituted >50% of the incidents. To their surprise, most incidents were not in the emergency department and psychiatry units, but in the medical/surgical units. In the last 2 fiscal years, 85% of their workplace violence incidents represented patient aggression to staff.
BIDMC needed to ensure that increased reporting would result in an improvement in workplace safety. They put in place an electronic safety alert flag for patients with demonstrated risk of violence to staff. This warns and protects staff from repeat harm. They instituted a Threat Assessment Response Team, consisting of Legal, Public Safety, Clinical, Health Care Quality, Social Work, and Patient Care. This team can be called to an area of immediate threat to support staff. They also reviewed and standardized hospital procedures for patient elopement to minimize the risk of violence associated with these events.
Finally, workplace safety education is standardized and accessible organization-wide. Standardized active workplace violence drills take place, not limited to Code Silver. Unit-based training occurs in both inpatient and ambulatory units. An e-module on deescalation is accessible for staff education.
With this 4-pronged strategic effort, staff at BIDMC can report and receive immediate support and response to a threatening workplace violence situation. The aggregated data—reported all the way up to the board—provide the hospital with the ability to learn and further reduce violent activity.
Virginia Mason is a health system whose CEO strongly messages their workplace violence prevention efforts. These efforts are part of an overarching strategic initiative and model of a culture of respect (Gary Kaplan, MD, CEO, personal oral communication, October 2018). In 2012, as part of their journey toward safer care, Virginia Mason embarked on training every member of their workforce in a program entitled Respect for People. This training includes trigger vignettes acted out by a local improvisation troupe, highlighting both respectful and disrespectful behaviors. Virginia Mason created a list of respectful behaviors. Every person at Virginia Mason, including the CEO, selects and commits to improvement in two out of ten listed behaviors: (1) listen to understand, (2) keep your promises, (3) be encouraging, (4) connect with others, (5) express gratitude, (6) share information, (7) speak up, (8) walk in their shoes, (9) grow and develop, and (10) be a team player.
In 2018, this program expanded to address nonphysical, verbal scenarios where patients might disrespect staff, staff might unintentionally disrespect patients, and staff might disrespect each other. This 2.0 training includes examples of “upstander” response to workplace violence—how a bystander can intervene in a critically productive manner.
Peer Support Systems
Part of the patient safety journey has included a focus on disclosure and the “care of the caregiver” after adverse events. These “peer support” programs can be resourceful as part of a supportive team response to workplace violence.
Jo Shapiro, MD, pioneered The Center for Professionalism and Peer Support at Brigham and Women’s Hospital in 2008. Its foundational base is that “what is paramount is the well-being of those who work here.” Her program’s16 framework includes the domains of professionalism, teamwork training/conflict management, just culture, disclosure coaching, and peer support. Knowing that 88% of those involved in an incident want to be supported by a peer colleague, the program has 50 to 60 trained peers. Although group peer support and Employee Assistance Programs (EAP) are available as additional resources, they learned that peers benefit from 1:1 peer support. These peer supporters are nominated for their relational skills and credibility. They complete a 5-hour training program, including simulation training. Importantly, this peer intervention is not about judgment, interpretation, or fixing the situation. The peer supporters develop and practice skills that emphasize support. The intervention includes the following components: outreach call, invitation/opening (can you tell me about what happened), listening, reflecting (“normalize” feeling traumatized), reframing, sense-making, coping, closing, and resources/referrals.
Dr Shapiro comments, “creating a peer support program is one way forward, away from a culture of invulnerability, isolation, and shame and toward a culture that truly values a sense of shared organizational responsibility for clinician well-being and patient safety.” Over 25 programs have been modeled nationally and internationally after this pioneering work.
Team and Self-education Resources
Organizations such as Crisis Prevention Institute (CPI)17 provide nonviolent crisis intervention and deescalation training. This is a valuable resource, especially with the survey results reported in this article showing that 75% of respondents had not received any deescalation training (Table 2). Verbal and nonverbal skills are distilled into 10 deescalation tips: (1) be empathetic and nonjudgmental, (2) use nonthreatening nonverbals, (3) focus on feelings, (4) set limits, (5) allow silence for reflection, (6) respect personal space, (7) avoid overreacting, (8) ignore challenging questions, (9) choose wisely what you insist upon, and (10) allow time for decisions.
When implementing workplace violence interventions, receiving data as a unit-based team seems key to success. In a randomized-controlled intervention, Dr Arnetz and her research team18,19 compared units who received unit-based workplace violence data [eg, # of incidents, type of incidents (patient to staff, staff to staff, role of those involved, worker loss time in costs)] to a control set of units receiving no data. After 6 months, the incident rate of violent events was ~50% lower in the intervention units compared with the control units (incident rate ratio, 0.48; 95% CI, 0.29-0.80). At 24 months, the violent injury rate was lower at ~60% compared with the control group (incident rate ratio, 0.37; 95% CI, 0.17-0.83). This suggests the importance of providing data-driven feedback at the unit level in optimizing results.
Anesthesiologists are not immune to physical and nonphysical workplace violence. In our anonymous survey, conducted in the fall of 2018, 20% of 2694 anesthesiologists reported physical workplace violence some time in their career, with patients and patients’ families being the most common aggressors. Sixty-nine percent of anesthesiologists experience nonphysical (eg, nonverbal/intimidation) workplace violence in their career, with physicians from another specialty being the most common aggressors, followed by patients and patients’ families. Most survey respondents did not feel that the incident was addressed and resolved to their satisfaction.
We need to address this important finding and develop action plans from leadership to the front-line bedside. Workplace violence reflects on workplace safety. Workplace safety reflects on patient safety. The following can be used as a mnemonic toward getting ready, developing, and implementing an action plan against workplace violence: I’M READY.
- I: Identify your own data on workplace violence and share data that are provided in this article.
- M: Map resources that are within the hospital (eg, peer support, security, employee assistance).
- R: Resources—supplement resources (eg, gap analysis) with statewide, national, and best practices.
- E: Engage with leadership—everyone should be treated with dignity and respect every day.
- A: Action—decide on how to deliver the message of workplace violence zero tolerance. This is NOT part of the job.
- D: Deescalation skills—improve your personal skill set of deescalation tactics. Role play with members of your workplace team.
- Y: Year—no more delays. Start this year.