Workplace violence in healthcare settings: Does preventive training and education affect prevalence?

Based on survey data, 66.01% of the participants have experienced patient-initiated violence within the past 6 months. Figure 1 illustrates the frequency of episodes as reported by participants. Participant data indicate that 167 participants experienced either physical abuse, verbal abuse, or both; of these, the majority experienced both forms of abuse (59%), while 39% reported verbal abuse only. The breakdown of types of verbal and physical abuse is displayed in Figures 2 and 3.

Figure 1.

Prevalence of patient-initiated violence

Figure 2.

Categories of physical abuse

Figure 3.

Categories of verbal abuse

Overall results showed that 73.76% of subjects were offered training on workplace violence by the facility in which they were employed. Face-to-face training was offered to 24% of respondents with 42% receiving online training only. A mix of face-to-face and online education was provided to 34% of participants. The chi-square test of independence was used to evaluate quantitative data to determine if patient violence and de-escalation training impacted the prevalence of patient violence experienced by healthcare providers. Statistical significance was set at alpha = 0.05 and data were analyzed using Intellectus Statistics, a web-based application.

The results of the Chi-square test were not significant based on an alpha value of .05, chi²(2) = 1.46, p = .483, suggesting that the two variables ‘Have you been involved in one or more episodes of violence’ and ‘Have you-Choice #1-completed the training at your place of work, Choice #2-Not completed this training, Choice #3-Completed the training at your own expense outside of your place of work’ could be independent of one another.

Cramér’s V was used to evaluate whether the participant received any training compared to the prevalence of patient-initiated violence experienced in the last 6 months. Results showed Cramér’s V to equal 0.0824, which is not significant on a 0-1 scale. Further review of the data indicated that although most participants completed training on how to manage aggressive and violent situations, training was not significantly associated with the prevalence of patient-initiated violence. In addition, Cramér’s V was used to evaluate whether training held face-to-face, online, or a blend of both had an impact on the prevalence of patient initiated violent episodes. Cramér’s V was shown to be insignificant at 0.110 on a 0-1 scale. However, as illustrated in Table 2, there was a slight decrease in violent episodes for those who had in-person or blended training, compared to online format only. Multiple participants responded in the survey qualitative comments that their training was neither frequent enough nor adequate to know how to effectively de-escalate aggressive or violent behaviors with patients.

Table 3 illustrates the reported number of violent episodes categorized by practice area. The largest group of participants were employed on a medical-surgical unit, and 83% of those participants on medical-surgical units reported experiencing episodes of violence within the last six months. Emergency Department and Behavioral Health units had a greater frequency of episodes reported. These results support the need for education and training on patient deescalation across healthcare facility units to manage episodes of patient violence.

Data from the study showed that approximately 50% of respondents felt that upper management was approachable following a violent event, where approximately 24% felt that upper management was not approachable, and 26% felt upper management was sometimes approachable. A greater percentage of respondents (68.66%) felt that managers/team leaders were approachable after a violent event. When respondents were asked if the risk management strategies to decrease workplace violence were effective, only 20.6% stated yes, whereas 54.27% stated somewhat and 25.13% said no. When respondents were asked about effective strategies, some commented that frequent and ongoing training was important, having a specifically trained management team to manage violence is important, and the organizations support for a safe workplace through the enforcement of a zero-tolerance culture. Other respondents stated they were not sure of what strategies had been implemented, if any. One participant commented: “No recent changes have been implemented despite physical attacks on staff.”

Most respondents (61%) felt that violence is increasing, and 84.58% of respondents answered yes (49.75%) or somewhat (34.83%) to the question that violence was part of the job. Respondent’s comments support the belief that violence is part of the job, and there is a strong need for organizational management of violence within healthcare. To the question “Is violence an inevitable part of your job” one participant commented: “We have created an environment where there is no consequence for when nurses and other healthcare personnel are abused making it okay and normal participant comments related to violence as part of the job, and the need for greater organizational management of violence when it happens.” Others stated: “I haven’t met a single nurse who hasn’t experienced violence.”, “Just part of the day to day in mental health care.”, “It’s just part of dealing with sick people”, and “It’s not acceptable, but I’ve learned to expect that it will happen at least once a week.” Another participant stated “I think it is inevitable that we will encounter violent patients or patients that are unhappy with the care they receive. However, in no way do I think it is acceptable or should be tolerated” which further supports the need for ongoing organizational recognition of the problem, and support for healthcare employees.

The qualitative questions on the survey explored healthcare workers’ perceptions on how to better manage patient violence in the workplace regardless of their level of experience with deescalation training. Responses to two questions participants were asked, “What measures do you believe can be used to more effectively to manage violent patients/episodes?” and “In your opinion, what would be the most effective way to prevent/minimize the occurrence of violence in your department?” strongly indicated that healthcare workers support frequent and on-going de-escalation training education. Table 4 is a sampling of several comments by participants that support the need for enhanced, frequent training within the workplace. Other participant comments (see Table 5) support the need for zero tolerance policies, better staffing ratios, and more overall support in the healthcare environment. More research into the most effective means of training and education for healthcare workers who experience violence in the workplace is needed.

Responses to the question, “Please add any additional comments below” provided a wealth of suggestions and recommendations for healthcare facilities in preparing workers to manage episodes of patient to healthcare worker violence. In addition, responses to this question support the need for more attention to patient violence in healthcare. One participant indicated: “It’s hard to say what would help. There seems like there will always be violence towards healthcare workers, especially verbal. Even though my department has lots of measures in place to help there doesn’t seem to be a method to truly stop it.” Another participant stated, “It is sometimes expected that our patients are not feeling their best and aggressive behavior is possible.” Further comments support the need for more attention to violence in the healthcare setting. “I love that you all are doing this project. Workplace violence against nurses is SO prevalent and should absolutely be taken more seriously.” “Thank you! Taking this survey has let me get things I have wanted to say for years off my chest.” “Please share your findings with as many people as possible. We get told we are complaining, and it’s what we signed up for if we say how we feel. The world needs to have a wakeup call to how we are treated and they won’t have any nurses left if the conditions continue.”

To further examine these relationships while accounting for potential confounding variables, an exploratory binary logistic regression was conducted to examine whether the availability of workplace violence training was associated with experiencing patient-initiated violence within the previous six months, while adjusting for years of experience, clinical setting, and employment status. The overall model was statistically significant, chi²(5) = 29.96, p < .001, indicating that the predictors, as a set, reliably distinguished between participants who reported experiencing violence and those who did not. The model demonstrated acceptable fit (Hosmer–Lemeshow test, p = .372) and explained approximately 17.3% of the variance in violence exposure (Nagelkerke R² = .173).

Availability of workplace violence training was not a statistically significant predictor of violence exposure (p = .912). Years of experience approached but did not reach statistical significance (p = .097). Clinical setting was a significant predictor (OR = 0.45, 95% CI [0.32, 0.64], p < .001), indicating that the likelihood of experiencing patient-initiated violence varied across practice environments. Employment status was not significantly associated with violence exposure (p = .383). These findings suggest that, after adjusting for key covariates, workplace violence training was not associated with reduced reports of patient-initiated violence within this sample.

This study has several important limitations. First, a notable dropout rate was observed in survey completion, which may have impacted the representativeness of the final sample. While the survey captured a wide range of data related to workplace violence, education, and participant perspectives, it did not include detailed measures assessing the effectiveness of specific training components (e.g., role-playing, simulation, or scenario-based learning), limiting the ability to evaluate which elements of training may be most beneficial.

The use of snowball sampling introduces potential selection bias and limits generalizability, as participants may share similar characteristics or professional networks. Additionally, data were self-reported and are subject to recall bias, as well as potential normalization of workplace violence experiences, which may have further influenced how participants interpreted and reported events.

The cross-sectional design precludes temporal sequencing and limits causal inference; therefore, findings should be interpreted as associative rather than causal. This is particularly important when considering the lack of a statistically significant association between workplace violence training and violence exposure, as unmeasured factors may contribute to this relationship. For example, clinical setting was found to be a significant predictor of violence exposure, suggesting that environmental and organizational factors may confound the relationship between training and outcomes.

Additionally, the modified survey instrument has not undergone formal psychometric evaluation beyond face validity, which may impact measurement reliability and construct validity. Finally, unmeasured confounding variables, such as workload intensity, staffing levels, and prior exposure to violence, may have influenced findings. Future research should consider longitudinal or pre-post designs that better assess the impact of workplace violence training over time, as well as more detailed measurement of training content and delivery methods to identify the most effective approaches.

Decreasing workplace violence involves a culture of change within healthcare organizations. One factor that is part of that change is the need for proper education for employees. Survey results indicated that a high percentage of respondents received either online, in-person, or blended education on workplace violence, with the majority completing online-based training. Results also indicated that patient violence was slightly higher for individuals who completed online training only. However, in the present study, no statistically significant association was observed between workplace violence training and prevalence of patient-initiated violence, even after adjusting for years of experience, clinical setting, and employment status. This finding suggests that training alone may not be sufficient to reduce the likelihood of experiencing workplace violence and highlights the need to consider additional contributing factors.

Previous research supports the need for education that involves role play and relatable scenarios. Providing in-person training allows time for hands-on practice and real-time feedback from educators. This is further supported by participant data from this research. A part of decreasing workplace violence is having education that is useful to the learner and applicable to their specific setting. Creating training that is setting-specific would allow the learner to practice in real-life scenarios that they may encounter, versus scenarios that allow them to engage in practice but are not applicable to their work environment.

Importantly, clinical setting emerged as a significant predictor of workplace violence exposure in this study, suggesting that environmental and organizational factors may play a more substantial role in violence risk than individual-level training alone. As noted in the research, many different units, both inpatient and outpatient, experience patient violence; however, the frequency and nature of that violence may vary considerably by practice environment. This reinforces the need for education that is not only broadly implemented but also tailored to the specific risks associated with different clinical settings.

In-person, scenario-based training should be required for all staff and all units, not just units that have a larger population of violence. This will allow staff to be proactive and prepared to manage violent situations when they occur. Consideration should also be given to the frequency of education. Qualitative data from this study supports the need for more frequent training within the hospital. Employees could complete a full training annually, followed by more frequent refresher sessions to maintain skill proficiency and confidence in managing escalating situations. In addition, healthcare organizations could conduct patient escalation drills, like fire drills, that are unannounced and allow the staff to practice skills in real time within their own units. Each drill should include structured debriefing to reinforce learning and identify areas for improvement.^[18]

The American Nurses Association^[19] discusses the need for early education of nurses, recommending de-escalation and crisis management training for nursing students prior to entering practice. This would allow nurses to enter practice prepared to manage patient violence and escalating situations. In addition, this would allow nursing students who are caring for patients in hospitals to manage potentially violent or escalating situations that arise during clinical rotations. Students could learn how to manage patients on a day-to-day basis by developing non-escalation skills, which focus on preventing escalation through therapeutic communication, appropriate behavioral responses, and awareness of environmental triggers.

Teaching only how to respond when escalation occurs represents a reactive approach, whereas teaching non-escalation techniques promotes prevention by emphasizing respectful, patient-centered interactions. Students should also learn how to respond when escalation occurs, including the use of verbal, non-verbal, and physical strategies to safely manage crisis situations.^[20] Providing this education would promote safety for both students and patients and support the delivery of safe, high-quality care.^[21]

A recommendation from study participants was the formation of Behavioral Emergency Response Teams (BERT), along with specific training for team members. In addition to security personnel, BERTs could include a multidisciplinary group of healthcare professionals such as physicians, nurses, and social workers who are trained to respond to escalating situations. The intent of the BERT team would not be to manage all patient escalation events, but rather to intervene early when staff recognize warning signs of escalation, potentially preventing progression to violence.^[19,22] These individuals could receive advanced training and serve as a resource to frontline staff, helping to promote both patient and staff safety. BERT teams could also lead unit-based de-escalation drills, further reinforcing preparedness across the organization.

Participant comments suggested the implementation and use of zero-tolerance policies in healthcare workplaces. A fully publicized and enforced zero-tolerance policy would ensure that patients are aware they will be held accountable for their actions toward healthcare workers.^[19,23] In addition, legislation in support of these policies, including laws that enforce consequences for individuals who harm healthcare workers, is essential. Healthcare facilities implementing clear policies, combined with supportive legislation, can create a unified approach to addressing workplace violence. Workplace violence can be understood within broader frameworks of organizational climate and psychological safety, which emphasize the role of leadership, culture, and systems-level factors in shaping employee experiences.^[24,25]

Importantly, the findings of this study suggest that organizational factors, including workplace culture, leadership responsiveness, and environmental context, may be critical determinants of workplace violence, particularly given that training alone was not associated with reduced violence exposure. Qualitative responses from participants further emphasized the need for stronger leadership engagement, consistent policy enforcement, and improved staffing patterns to effectively address workplace violence. Together, these findings reinforce that meaningful reductions in workplace violence will likely require multifaceted system-level interventions rather than reliance on education alone.