In the conclusion of our report on healthcare violence from the December 2016 issue of Hospital Employee Health, we look at some underlying causes and much-needed solutions in a conversation with officials in Boston, which has suffered healthcare violence and a terrorist attack in recent years.

An incident that shook the Boston medical community occurred in January 2015, when a surgeon at Brigham and Women’s Hospital was shot and killed at work by a relative of a deceased patient. That shocking event, on the heels of the 2013 Boston Marathon bombing, has certainly instilled a sense of readiness and vigilance in the city’s healthcare facilities.

That said, the open campus of a medical facility cannot be locked down like airport security, so officials strive for a balance between delivering care and protecting patients and healthcare workers.

“The question always comes up: Do you have metal detectors? Do you arm your staff?” says Constance L. Packard, CHPA, executive director of support services at Boston University Medical Center.

She answers “no” to both questions, though her team has handheld wands for weapon detection if needed.

“You have to weigh what is most appropriate for patient care and keeping employees safe,” Packard says. “I learned a lot from my colleagues in the Boston Marathon bombing — have those tools and be able to put them in place when you need to use them. That’s versus having, for example, a stand-alone metal detector. We see 12,000 to 15,000 people here a day. It’s a very busy place with 42 buildings and hundreds of entrances. It’s just impossible to do that and do it well.”

Having developed active shooter training videos and scenarios, Packard praised The Joint Commission’s recent development of a violence prevention resources portal for healthcare. (To see the portal, visit:

“I know from going through Joint Commission accreditation for numerous years that emergency management and workplace violence are probably two areas that they do focus on when they look at environment of care standards,” she says. “They are interested in how you are doing your program, and more importantly, how you are educating your staff should an event occur.”

At Boston Medical, that education includes an ongoing reminder to staff that no incident is “too little” to report.

“We want them to give us a heads-up that [for example, a patient] has an outburst,” Packard says. “It gives us, from a public health perspective, the ability to look at risk. As police officers, it gives us the opportunity to reduce vulnerabilities. In the last year, we have worked with our IT department to put something in the [patient] chart called administrative precautions — an FYI or flag when we have a problem patient or a problem family member.”

The staff are supportive of having that information in the patient chart, which may encourage reporting of incidents that may otherwise go untracked.

“I can tell you the staff has been very vocal about how important it is to see that information,” Packard says. “They have to call us and tell us about the event, whether it was the patient or the patient’s family, and then we will determine if it is at the level that needs to be [recorded in the medical record]. It may just generate a public safety report or it may develop what we call a comprehensive care plan, but we have nothing to lose by sharing the information.”

While such policies may encourage more reporting of violent or threatening incidents, many healthcare workers are reluctant to take it a step further and formally press charges against patients they may regard as suffering from drug-related behavior or mental health problems.

“Others, depending on the injuries, say, ‘Yes, I want to go forward,’ and we work with them and pursue prosecution in the courts,” Packard says.

Though some violent incidents are completely unpredictable, the training advised by Packard and others can still mitigate a situation and save additional injuries or loss of life. In July 2016, a man entered Parrish Medical Center in Titusville, FL, and fatally shot a patient and caregiver for no apparent reason. Security guards managed to disarm the man and hold him for law enforcement, prompting this comment from Titusville Police Chief John Lau: “I cannot stress enough [that] the response of the Parrish Medical Center staff, without a doubt, saved more lives.”1

The Broken Window

The failure to report and act when an incident may be verbal or seem minor can contribute to the “broken window” effect, which essentially suggests that an ongoing tolerance for a low level of crime may contribute to its subsequent escalation, says James P. Phillips, MD, of the Department of Emergency Medicine at Beth Israel Deaconess Medical Center in Boston.

“Both verbal and nonverbal violence from patients against providers are classified as ‘Type II’ workplace violence,” Phillips explains to Hospital Employee Health. “The broken windows theory is adapted from street crime, and [suggests] that intolerance of low-level crimes such as broken windows — or in the case of workplace violence, verbal escalation and disrespect — helps to prevent higher-level crimes such as physical battery in the workplace.”

Phillips wrote a definitive review article2 on the problem of violence in healthcare last year in the New England Journal of Medicine, noting that while the murder of a healthcare worker certainly draws national attention, there is still an underappreciation of the scale of the daily problem.

“[E]pisodes of workplace violence against medical providers happen daily across the country,” Phillips wrote in the review article. “Although the majority of these incidents of workplace violence are verbal, many others constitute assault, battery, domestic violence, stalking, or sexual harassment.”

In the paper, Phillips places violence into four categories, including the aforementioned Type II. Each of the categories reflect the relationship of the perpetrator to the workplace or employees:

  • Type I: No association (e.g., person with criminal intent commits armed robbery).
  • Type II: Perpetrator is a patient or customer of the workplace or employees (e.g., intoxicated patient punches nurse’s aide).
  • Type III: A current or former employee of the workplace (e.g. recently fired employee assaults former supervisor).
  • Type IV: Perpetrator has a personal relationship with employees, none with the workplace (Ex-husband assaults ex-wife at her place of work).

As Phillips emphasizes, Type II violence — which can be physical or verbal assaults — is the most common form faced by healthcare workers. Indeed, a 2014 study3 found that healthcare workers had the highest number of Type II incidents in U.S. workplaces. Type II workplace violence accounted for 75% of aggravated assaults and 93% of all assaults against employees, Phillips reports.

In addition, there appears to be a direct relationship between patient contact time and possible violence, placing nursing aides and nurses at the greatest risk. Not surprisingly, ED nurses report high rates of violent incidents, and in one study4 approximately 25% of emergency medicine physicians reported being targets of physical assault in the previous year.

Given the level of violence against healthcare workers documented, it is disconcerting to consider Phillips’ conclusion that episodes of workplace violence of all categories are “grossly underreported.” In particular, he notes that nurses have cited fear of retribution from supervisors and disapproval of administrators as barriers to reporting, possibly in part due to the prevailing “the customer is always right” mentality. This raises the disturbing connotation that somehow a level of violence is ingrained in the healthcare work culture.

“The recent trend over the last two decades toward viewing a patient as a customer has had deleterious effects on the patient-physician and patient-nurse relationships in many ways, in my opinion,” Phillips says. “There has been a depersonalization, and a subsequent decrease in respect between patients and physicians in both directions. This obviously is not the case in every field or every situation, but I think most physicians and nurses would agree with my opinion in general. That decrease in respect for the provider has certainly contributed to increased verbal disrespect and probably violence.”

By the same token, the patient as a customer expecting quality service may be less tolerable to long waits in the ED, beds temporarily placed in hallways, or lack of nursing and physician attention due to census overload.

“Patients who do not feel they are getting their ‘money’s worth’ are much more likely to act out, in my opinion,” Phillips says. “In regards to underreporting, there are many barriers that prevent victims from documenting and reporting such incidents. There is a general feeling that there is no worthwhile consequence or punishment for the offender that is worth the time and energy of the provider-victim to pause their work, fill out a form, make a phone call, or file a police report.”

Healthcare administration must address this problem, encouraging reporting and taking action to overcome this depressing status quo.

“Providers have not been made to feel that violence prevention and redress is a priority in their workplace despite the statistically proven risk in healthcare, and without that, I do not think we will see much increase in reporting.” Phillips says. “It is an administrative responsibility to acknowledge that this violence exists in every healthcare setting, and that if individual institutions do not start making real improvements, the government and accrediting bodies will eventually regulate them into doing so.”

Budgeting for Violence Prevention?

Phillips and colleagues are undertaking a multiple hospital study to look at whether facilities have sufficiently dedicated budgets and policies to protect healthcare workers from violence.

“Preventing workplace violence is not free — it may not be cheap,” he says. “It certainly requires a budget, and employees who are working in the field should not be expected to volunteer their time to fix this problem. I would bet that the large-scale hospital study we are beginning now will demonstrate that most do not have a dedicated budget for prevention of patient-to-provider violence, and I hypothesize that many probably do not have any hospital policy on the subject, either. Equipment and personnel are expenses, and for administrators who are responsible for keeping the lights on — and for some, generating profits — I do not think we will see widespread change until OSHA, The Joint Commission, or government regulations mandate it.”

In the interim, employee health professionals and their healthcare colleagues can take the first step by acknowledging the problem and raising awareness.

“Each hospital must first admit that they are not special, and that such violence occurs in every facility,” Phillips says. “I understand the apprehension that hospitals may have toward publicly acknowledging that there is violence in their facility. Even more so, it will take a brave, dedicated administration to be willing to allow research and release of violence statistics in their hospital because of the perceived risk that such public admission will be deleterious to patient and employee recruitment. I ask for such courage from hospital administrations.”

Phillips suggests establishing a multidisciplinary workplace violence committee that includes administrators, supervisors, technicians, nurses, and physicians and other employees. With the input of clinicians and providers, develop a basic reporting system that includes — as cited by Packard — a “red flag” warning to future providers of a patient or family member with a history of violence.

“Most importantly, frontline providers must be made certain that their administrators and supervisors will support them fully and will ‘have their backs’ in such cases,” Phillips says. “This requires real communication and acknowledgment of the issue. Hospitals should eliminate customer service measurements like the Press Ganey scale, which are often tied to physician payment. [These scales hinder] the provider setting limits with patients when disagreements and conflict arise, which ultimately contributes to the continued decline in the quality of the patient-physician relationship.”


  1. Ferenc, J. Florida hospital proves an active shooter plan can save lives. American Hospital Association: Healthcare Facilities Management. August 18, 2016:
  2. Phillips, JP. Workplace Violence against Health Care Workers in the United States. N Engl J Med 2016; 374:1661-1669.
  3. Vellani KH. The 2014 IHSSF crime survey. J Healthc Prot Manage 2014;30:28-35.
  4. Kowalenko T, Gates D, Gillespie GL, et al. Prospective study of violence against ED workers. Am J Emerg Med 2013; 31:197-205.