Introduction

Workplace violence (WPV) refers to incidents in which an individual is exposed to any type of violence in the workplace; it can be in the form of physical, verbal, emotional, or sexual violence (Nowrouzi-Kia et al. 2019). This observation is variable based on the type of workplace. In particular, hospital settings have been reported to have one of the highest incidents of WPV against healthcare workers (HCWs) (Fute et al. 2015). In the USA, according to the Occupational Safety and Health Administration (OSHA), healthcare settings are more likely to experience and report serious WPV as compared to private industry(OSHA 2015).

WPV has occurred in epidemics and pandemics throughout the past century. From the 1830s to the 1910s, cholera infection control and public health workers encountered suspicion and antagonism (Cohn and Kutalek 2016). Those enforcing mask regulations during the 1918–1920 H1N1 epidemic resulted in violence against those who enforced mask-wearing (Dolan 2020). Healthcare professionals, journalists, and government officials were attacked and threatened during the 2014–2016 Ebola pandemic (Cohn and Kutalek 2016). Thus, the present pandemic mirrored history.

WPV is not a recent phenomenon. Previous reports highlighted the prevalence of WPV against HCWs of various types, either nurses, physicians, or other workers in the hospital setting (Liu et al. 2019a). In 2021, a meta-analysis reported that the rates of any type of WPV, verbal abuse, physical violence, sexual harassment, and bullying among nurses, particularly in South-East Asian and Western Pacific countries, are 51–64%, 59–70%, 14–34%, 7–17%, and 17–33%, respectively (Varghese et al. 2022). Gender, occupation, practice settings, and work schedules have all been linked to WPV (Gillespie et al. 2017; Fujita et al. 2012). These variables may differ between countries.

The COVID-19 pandemic has resulted in a significant impact on healthcare workers in terms of reduced sleep (Shreffler et al. 2020), increased workload (Teo et al. 2021), and increased risk of morbidity and mortality associated with the disease (Bandyopadhyay et al. 2020), besides the substantial negative impact on their mental health (Van Wert et al. 2022). Additionally, the COVID-19 pandemic has inferred an extra burden on HCWs, which subsequently increased the rates of burnout (Ferry et al. 2021) and affected the quality of provided care (Baskin and Bartlett 2021) and their tendency to leave their jobs (Rose et al. 2021).

The implementation of methods for avoiding and managing the spread of Sars-COV-2 infection, such as quarantine and isolation, increases the likelihood of violent or threatening behavior by patients and families (Yang et al. 2021a). Workplace violence against healthcare personnel is related to poor outcomes such as dissatisfaction with work, healthcare errors, psychological issues, and decreased service quality, and is thus a substantial public health issue (Heponiemi et al. 2014; Bellizzi et al. 2021; Aljohani et al. 2021). This is becoming a more serious problem around the world, and it appears to impact ED professionals more frequently and profoundly than in other clinical settings (Cooper and Swanson 2002). This could be ascribed to overcrowding and extended wait times, as well as individual perpetrator characteristics such as alcohol, drugs, or mental disorders (Timmins et al. 2023).

A recent systematic review of 17 studies highlighted the prevalence of WPV among HCWs which was estimated at 47%, where psychological violence (44%) was higher than physical one (17%) (Ramzi et al. 2022). The authors also highlighted that the rate of WPV was higher among physicians compared to other HCWs. That being said, since then, a large number of studies have been published. Therefore, we conducted this updated systematic review and meta-analysis to determine the rate of different types of WPV against HCWs with emphasis on the variability in WPV according to the profession, medical specialty, and the timing of the COVID-19 pandemic. Additionally, we aimed to identify the risk factors associated with WPV among HCWs.

Materials and methods

Search strategy and setting

This research was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, where the pre-registration of a protocol is not mandated. The design of this research followed the PICOS framework as follows: population (healthcare workers of any type: physicians, nurses, others), intervention (none), comparison (none), outcomes (the prevalence of various types of workplace violence in hospital settings during the COVID-19 pandemic), and study design (observational studies).

On May 16, 2022, PubMed, Scopus, Web of Science (WoS), EBCOHost – Academic Search Complete, and Google Scholar were searched for studies reporting the occurrence of WPV against HCWs during the COVID-19 pandemic. Noteworthy, based on recent recommendations (Muka et al. 2020), only the first 200 records of Google Scholar were searched, after which relevance significantly dropped. The following keywords were used to identify relevant articles: (“healthcare worker” OR physician OR nurse) AND (“workplace violence” OR bully OR harassment) AND (COVID-19 OR SARS-CoV-2). Whenever possible, Medical Subject Headings (MeSH) terms were used to identify all potentially relevant articles. The search criteria were then adjusted based on the selected database. A full description of the search query used in each database is provided in Supplementary Table 1. Additionally, a manual search was also conducted following the screening of articles to identify any potentially missing relevant article through three approaches: (a) screening the reference list of included articles, (b) screening “similar articles” to included ones through the “similar articles” options on PubMed, and (c) manually searching for articles on Google with the use of following keywords: “COVID” + “healthcare worker” + “workplace violence”. Noteworthy, an updated search was carried out on Oct 20, 2022, to include any newly published studies prior to the analysis.

Sample

Studies were included if they were compliant with all of the following criteria:

  • Observational studies that were conducted during the COVID-19 pandemic.

  • Including healthcare workers as their target population.

  • Reporting the rate of workplace violence towards HCWs.

On the other hand, studies were excluded if they had one of the following criteria:

  • Experimental studies and non-original research (reviews, editorials, letters, commentaries, etc.).

  • Studies conducted before the beginning of the pandemic.

  • Studies reporting WPV against individuals other than HCWs.

  • Studies reporting irrelevant outcomes.

  • Qualitative studies.

  • Duplicated records or studies with overlapping datasets.

Following the retrieval of studies from the database search, citations were imported into EndNote for duplicate removal, after which, citations were exported into an Excel Sheet for screening. First, the titles and abstracts of retrieved articles were screened against our prespecified eligibility criteria. Then, studies that were potentially relevant underwent full-text screenings. Noteworthy, the selection of studies was not dependent on the type of HCW, WPV, or hospital setting. Additionally, studies were included regardless of who the perpetrator was. This process was carried out by two reviewers who solved their differences through discussions. Meanwhile, the senior author was consulted when an agreement could not be reached.

Data extraction

A pilot extraction was carried out to design the data extraction sheet using Microsoft Excel. The data extraction sheet consisted of two parts. The first part included the baseline characteristics of included studies (first author’s name, year of publication, country, study design, timing of COVID-19, and follow-up period) and patients (age, gender, profession, specialty, and care for COVID-19 patients). The second part included the study outcomes (overall workplace violence and physical, verbal, combined, emotional, and sexual violence/abuse in addition to harassment and bullying). Combined violence referred to cases that reported being exposed to different types of violence simultaneously, including physical, verbal, sexual, and emotional. For reporting purposes, the use of the word “WPV” will refer to violence of any type. Two reviewers extracted the relevant data from included studies, and any discrepancies, inconsistencies, or data entry mistakes were revised through group meetings with the senior author.

The primary outcome of this meta-analysis is to estimate the global prevalence of WPV against HCWs during the COVID-19 stratified by the type of violence (physical, verbal, combined physical and verbal, emotional, sexual, harassment, bullying), the type of HCWs (physician vs. nurse), the stage of the pandemic (early or 2020 vs. mid or 2021 vs. late 2022), and specialty (pediatrics, internal medicine, neurosurgery, emergency medicine, etc.).

The secondary outcomes included the identification of risk factors of WPV (overall) and its two main categories (physical and verbal) separately. The considered risk factors were gender (male vs. female), care for COVID-19 patients (yes vs. no), the timing of the pandemic (before vs. during), and profession (nurse vs. physician).

The quality of included studies (cross-sectional in design) was assessed using the Newcastle Ottawa Scale (NOS) which assesses the quality of each study at the level of three domains: selection (4 questions), comparability (2 questions), and outcome (2 questions). Each study was given an overall score from 0 to 10, based on which the quality was determined as follows: good (3 or 4 points in the selection domain + 1 or 2 points in the comparability domain + 2 or 3 points in the outcome domain), fair (2 points in the selection domain + 1 or 2 points in the comparability domain + 2 or 3 points in the outcome domain), or poor (0 or 1 point in the selection domain OR 0 points in the comparability domain OR 0 or 1 point in the outcome domain). The quality was assessed by two reviewers each and any differences between them were referred directly to the senior author for a final decision.

Data analysis procedure

All statistical analyses were carried out through STATA Software (Version 17). The metaprop command was used to pool the prevalence rate — effect size (ES) — and its corresponding 95% confidence interval (CI) across studies. The random-effects and fixed-effects models were selected according to the presence or absence of heterogeneity, respectively. Heterogeneity was present if the I2 statistic was above 50% and the P-value was < 0.05. A subgroup analysis was then conducted based on the type of WPV, type of HCW, type of medical specialty, and timing of the COVID-19 pandemic. Regarding secondary outcomes, the restricted maximum likelihood method (REML) and the Mantel–Haenszel method were used if heterogeneity was present or absent, respectively. Then, the log of the odds ratio (logOR) and its corresponding 95% CI were reported. A leave-one-out sensitivity analysis was carried out to determine if the reported effect estimate of each outcome was driven by a particular study. Of note, all analyses were conducted per participant not per incident (violence).

Results

Search results

The database search and screening processes are illustrated in Fig. 1. The initial database search resulted in 1797 records, of which 414 duplicates were removed through EndNote. Then, the titles and abstracts of 1383 articles were screened against our eligibility criteria, out of which 51 were eligible for full-text screening. Thirty-one articles were finally deemed eligible for inclusion, while 20 were excluded for the following reasons: irrelevant outcome (n = 11), qualitative study (n = 3), unextractable data (n = 2), studies reporting WPV incidents per visit not per HCW (n = 3), and data included HCWs’ family members (n = 1). The manual search yielded two additional studies and the updated database search resulted in five more studies. Overall, 38 studies were eligible for both the qualitative and quantitative analysis of our review (Abdelhafiz et al. 2020; Alameddine et al. 2021; Alfuqaha et al. 2022; Alves et al. 2022; Angwenyi et al. 2021; Asaoka et al. 2021; Aspera-Campos et al. 2020; Bitencourt et al. 2021; Buran and Altın, 2021; Byon et al. 2021; Chowdhury et al. 2022, 2021; De la Cerda-Vargas et al. 2022; Dopelt et al. 2022; El Ghaziri et al. 2022; García-Zamora et al. 2022; Garg et al. 2020; Ghareeb et al. 2021; Guo et al. 2022; Iida et al. 2022; Kashtanov et al. 2022; Khatatbeh et al. 2021; Kuhlmann et al. 2021; Kurzthaler et al. 2021; Lafta et al. 2021; McGuire et al. 2022; Mediavilla et al. 2021; Muñoz Del Carpio-Toia et al. 2021; Özkan Şat et al. 2021; Qi et al. 2022; Sarfraz et al. 2022; Serafin et al. 2022; Somville et al. 2021; Xie et al. 2021; Yang et al. 2021a, b, 2022; Zhizhong et al. 2020).

Fig. 1
figure 1

The PRISMA flow diagram of the database search and screening processes

Full size image

Study characteristics

The baseline characteristics of included studies are presented in Table 1. All studies were cross-sectional in design. One study was conducted in Austria, two in Bangladesh, one in Belgium, two in Brazil, seven in China, one in Ecuador, one in Egypt, one in Germany, one in India, one in Iraq, one in Israel, two in Japan, three in Jordan, one in Kenya, two in Latin America, one in Lebanon, one in Mexico, one in Peru, one in Poland, one in Russia, one in Spain, two in Turkey, and three in the USA. The sample size of individual studies ranged from 67 (Buran and Altın, 2021) to 15,531 (Yang et al. 2021a) HCWs, with an overall number of 63,672 HCWs being analyzed. The majority of participants were females, while the male gender accounted for 22.57% of them (12,775 out of 56,581 HCWs). The percentage of HCWs who cared for COVID-19 patients ranged from 0.63% (El Ghaziri et al. 2022) to as high as 100% (Abdelhafiz et al. 2020; Buran and Altın, 2021). Nine studies included nurses only, 10 included physicians only, 18 included mixed HCWs, and one study did not clarify the type of HCW.

Table 1 The baseline characteristics of included studies
Full size table

The quality of study in each of the examined domains is reported in Supplementary Table 2. Overall, five studies showed fair quality, eight studies had poor quality, and the remaining had good quality.

Outcomes

The prevalence rate of each type of violence against HCWs is presented in Table 2. Our meta-analysis of 26 studies (50,077 HCWs) revealed an overall prevalence of WPV of 43% (95% CI: 34–51%). The rate of verbal violence was the highest (48%; 95% CI: 48–48%), followed by emotional (26%; 95% CI: 8–45%), physical (9%; 95% CI: 9–10%), and combined violence (4%; 95% CI: 3–5%). Harassment, of any type, was reported by 7% (95% CI: 5–8%) of HCWs, while the rate of sexual harassment, in particular, was higher (8%; 95% CI: 7–9%). Among four studies (1087 HCWs), 27% (95% CI: 24–29%)] reported being bullied.

Table 2 Quality of included studies using the Newcastle Ottawa Scale for cross-sectional studies
Full size table

The detailed prevalence rate of each type of violence, stratified by the stage of the COVID-19 pandemic, is provided in Tables 3, 4, 5, and 6. The rate of WPV was highest during the early-pandemic (61%) as compared to the mid-pandemic (40%) and late-pandemic periods (47%). Of note, the rate of physical violence during the late-pandemic period (23%) was almost double that of the mid-pandemic period (12%); however, the difference in the rate of verbal violence was as much during the late- and mid-pandemic periods (58% vs. 45%), respectively. Surprisingly, the rate of emotional violence was lower during the late-pandemic (5%) as compared to the mid-pandemic (28%) period.

Table 3 The prevalence of workplace violence against healthcare workers stratified by the type of violence
Full size table
Table 4 The prevalence of workplace violence against healthcare workers stratified by the stage of the COVID-19 pandemic
Full size table
Table 5 The prevalence of workplace violence against healthcare workers stratified by the type of healthcare worker
Full size table
Table 6 The prevalence of workplace violence against healthcare workers stratified by the type of medical specialty
Full size table

The detailed prevalence rate of each type of violence, stratified by the type of healthcare worker, is provided in Supplementary Table 3. Overall, the rate of WPV between nurses (41%) and physicians (48%) was quite similar, which is consistent with the rate of verbal violence between nurses and physicians (56% vs. 57%), respectively. That being said, the rate of physical violence against nurses (13%) was more than double that of physicians (5%). On the other hand, the rate of emotional violence is remarkedly lower among physicians (3%) as compared to nurses (27%).

The prevalence of WPV against HCWs stratified by specialty

The detailed prevalence rate of each type of violence, stratified by the type of medical specialty, is provided in Supplementary Table 4. Unfortunately, no insightful evidence can be deducted from this analysis since the rate of different types of violence in certain specialties was dependent on the analysis of one-to-three studies. However, it should be noted the rate of WPV in Psychiatry was the lowest (3 studies, 13%), followed by Emergency Medicine (3 studies, 34%) and Internal Medicine (41%) department, respectively.

Based on the meta-analysis of 10 studies, no significant difference in the risk of experiencing WPV was determined based on gender (LogOR = 0.01; 95% CI: − 0.21: 0.23; I2 = 92.06%) (Fig. 2). The leave-one-out sensitivity analysis did not reveal any significant change in the reported effect estimate following the removal of one study at a time (Supplementary Fig. 1).

Fig. 2
figure 2

A forest plot showing the risk of workplace violence against healthcare workers based on gender

Full size image

Consistently, no significant difference in the risk of both physical (LogOR = 0.43; 95% CI: − 0.10: 0.96; I2 = 87.73%) (Fig. 3) and verbal (LogOR =  − 0.32; 95% CI: − 1.17: 0.53; I2 = 93.41%) violence (Fig. 4) was noted among male and female HCWs. Also, the leave-one-out sensitivity analysis did not reveal any significant change in the reported effect estimates regarding physical (Supplementary Fig. 2) and verbal violence (Supplementary Fig. 3).

Fig. 3
figure 3

A forest plot showing the risk of physical violence against healthcare workers based on gender

Full size image
Fig. 4
figure 4

A forest plot showing the risk of verbal violence against healthcare workers based on gender

Full size image

Based on the meta-analysis of 5 studies, no significant difference in the risk of experiencing WPV was determined based on profession (LogOR = 0.09; 95% CI: − 0.25: 0.43; I2 = 82.93%) (Fig. 5). The leave-one-out sensitivity analysis did not reveal any significant change in the reported effect estimate following the removal of one study at a time (Supplementary Fig. 4).

Fig. 5
figure 5

A forest plot showing the risk of workplace violence against healthcare workers based on the type of healthcare worker

Full size image

Consistently, no significant difference in the risk of both physical (LogOR = 0.07; 95% CI: − 1.17: 1.31; I2 = 93.64%) (Supplementary Fig. 5) and verbal (LogOR =  − 0.13; 95% CI: − 0.36: 0.10; I2 = 0%) violence (Supplementary Fig. 6) was noted between nurses and physicians. Also, the leave-one-out sensitivity analysis did not reveal any significant change in the reported effect estimates regarding physical (Supplementary Fig. 7) and verbal violence (Supplementary Fig. 8).

Based on the meta-analysis of 3 studies, HCWs who cared for COVID-19 patients were at higher risk of experiencing WPV (LogOR = 0.54; 95% CI: 0.10: 0.97; I2 = 95.92%) (Supplementary Fig. 9). The leave-one-out sensitivity analysis deemed this observation insignificant following the removal of the study of Xie et al. (2021) and Yang et al. (2021a) (Supplementary Fig. 10).

The risk of physical violence was significantly higher among HCWs who cared for COVID-19 patients as compared to those who did not (LogOR = 0.80; 95% CI: 0.39: 1.22; I2 = 71.23%) (Supplementary Fig. 11). The leave-one-out sensitivity analysis did not result in any significant change in the reported effect estimate (Supplementary Fig. 12). Meanwhile, no significant difference was noted in terms of verbal violence (LogOR = 1.88; 95% CI: − 0.72: 4.48; I2 = 99.26%) (Supplementary Fig. 13). Since this meta-analysis was based on two studies, the findings from the sensitivity analysis are not significant (Supplementary Fig. 14).

No significant difference was noted in the risk of WPV between during and before the COVID-19 pandemic (LogOR = 0.44; 95% CI: − 0.85: 1.74; I2 = 75.85%) (Supplementary Fig. 15). This analysis was based on only two studies; therefore, the findings from the sensitivity analysis are inconclusive (Supplementary Fig. 16).

Similarly, no significant difference was noted in terms of physical violence (LogOR = 0.29; 95% CI: − 0.56: 1.15; I2 = 75.85%) (Supplementary Fig. 17). This analysis was based on only two studies; therefore, the findings from the sensitivity analysis are inconclusive (Supplementary Fig. 18). Noteworthy, no studies reported the difference in risk of verbal violence during and before the pandemic.

Discussion

Our meta-analysis, of 38 studies and 63,672 HCWs, indicated around 43 out of 100 HCWs experienced some sort of violence during the COVID-19 pandemic. This rate was relatively higher in terms of verbal violence (48 out of 100 HCWs) while somewhat high regarding physical violence (9 out of 100 HCWs). These findings are of high certainty given the absence of heterogeneity and the very narrow confidence interval. A small number of studies reported a relatively low rate of sexual harassment (8%). Meanwhile, bullying occurred in more than one-fourth of the population (27 out of 100). These findings are similar to a previous meta-analysis that was published during the COVID-19 pandemic which reported that 47% and 17% of HCWs experienced any type of violence and physical violence, respectively (Ramzi et al. 2022). Furthermore, the monthly assault rate on total visits increased from 13.5 (SD 6.6) in the pre-COVID-19 era to 27.2 (SD 9.8) during the pandemic months in Italy (Brigo et al. 2022). This could be explained by the increased number of patients which led to the medical staff's stress and thus dissatisfaction of the patients and their families.

Based on the findings of a meta-analysis that was conducted before the beginning of the COVID-19 pandemic, our findings are somewhat comparable in terms of any type of WPV (61.9%), physical violence (24.4%), verbal violence (57.6%), and sexual harassment (12.4%) (Li et al. 2020). This is supported by our meta-analysis which highlighted no significant increase in the risk of both WPV (logOR = 0.44; 95% CI: − 0.86: 1.74) and physical violence (logOR = 0.29; 95% CI: − 0.56: 1.15) during the COVID-19 pandemic as compared to before. However, it should be noted that this finding is based on the analysis of only two studies; more studies are yet needed to confirm this observation.

The previous work of Ramzi and colleagues (2022) highlighted that the prevalence of any type of WPV was remarkedly higher among physicians as compared to nurses (68% vs. 47%), respectively. However, our meta-analysis highlights comparable rates of WPV (48% vs. 41%) and verbal violence (57% vs. 56%) between physicians and nurses, respectively. It should be noted, however, that the rate of physical violence among nurses was more than twice that of physicians. The causes behind this growth, particularly among nurses, have yet to be explored. Furthermore, the rate of emotional abuse/violence was significantly greater among nurses (27% vs. 3%). This discovery could be explained by nurses’ involvement and recognition of situations (such as the COVID-19 pandemic) that necessitate developing an emotional connection with patients and their families, putting them at a higher risk of encountering this sort of violence. (Jiménez-Herrera et al. 2020).

Our findings highlight the increase in the rate of WPV, physical, and verbal violence as the pandemic progressed from its peak (in 2021, mid-pandemic period) to the period of vaccination (in 2022). Remarkably, the rate of physical violence almost doubled between the mid-pandemic and late-pandemic periods (12% vs. 23%). Data on the differences in WPV rates according to the type of medical specialty or field are inconclusive due to the low number of studies with relevant data.

On the contrary, the Portuguese Association for Hospital Development reported that violent episodes have decreased from 9 per 1000 workers to 4 per 1000 workers in the year since the pandemic began, using data from a 14-year online reporting system. A 24-h remote video contact hotline with qualified specialists and a security focal point for personnel to coordinate with security officials helped reduce this (Thornton 2022).

After lobbying by nurses, Italy’s parliament passed a new law in 2020 to address violence against health workers, increasing prison time from 4 to 16 years for those who inflict serious or very serious personal injuries on health workers and increasing the administrative penalty from €500 to €5000 for actions that, while not crimes, violate health workers’ rights. The National Day of Education and Prevention of Violence against Health Personnel (March 12th) was created to raise awareness (Thornton 2022).

We noted no significant increase in the risk of WPV, physical, or verbal violence based on gender or the type of HCW (nurse vs. physician). This finding was in accordance with the previous study of Liu et al. who reported that gender did not affect WPV, non-physical, or physical violence prevalence (Liu et al. 2019b). When they examined gender differences in WPV prevalence in stratified settings within studies, they found that women were less likely to be exposed to physical violence in primary care (OR = 0.52; 95% CI: 0.29–0.92) and general hospitals (OR = 0.65; 95% CI: 0.47–0.89), but more likely to be sexually harassed (OR = 3.92; 95% CI: 2.70–5.70) (Liu et al. 2019b). This is inconsistent with the literature which reveals an increased risk of WPV in women as compared to men (OR = 2.12; 95% CI: 1.52–2.95) (Lanthier et al. 2018).

Al-Azzam et al. discovered that marital status, work duration, antiviolence strategy, and violence prevention training were all relevant predictors of WPV for mental health department nurses (Al-Azzam et al. 2018). Gender and employment settings (urban) were reported to be strongly linked with WPV in healthcare institutions by Jatic et al. (2019).

This difference, although existent, might differ based on the workplace setting; the previous study recruited participants from the general population (Lanthier et al. 2018). Moreover, our analysis revealed that HCWs who cared for COVID-19 patients were at higher risk of experiencing WPV (of any type) or physical violence. However, no significant difference was noted in the risk of verbal violence among those who cared for COVID-19 patients as compared to those who did not.

Overall, this study strongly encourages violence reduction and the safeguarding of medical places and workers safety. This could be achieved through deterrent government laws, unemployment reduction, media awareness, development of communication strategies between the medical staff and the patients, increase in the number of the medical staff, worker training, and education to face those situations and raise public awareness (Xiao et al. 2022; Coccia 2021, 2022; Marsh et al. 2022; Elsaid et al. 2022; Bellitto and Coccia 2018; Özdamar Ünal et al. 2022).

Study limitations and future directions

Our study highlights the magnitude of the problem of WPV against HCWs during the COVID-19 pandemic as well as the need to develop well-structured reporting systems in each hospital setting while encouraging HCWs to report such incidents in a timely manner. In addition, our findings urge the need to implement readily available psychological support systems for HCWs. That being said, there are still some gaps that are worth investigating, which we did not assess in our study, in order to provide a full picture of this phenomenon.

First, our review shows the prevalence of violence against HCWs, but we did not consider the number of violent incidents per HCW, which is important to consider in future studies because, presumably, the more an HCW is exposed to violence, the more it will affect their performance and desire to work. Second, future research should focus on hospital perpetrators (patients, family members, coworkers, etc.) to provide additional evidence for effective prevention. Third, data on HCW violence in different medical professions is sparse; thus, more studies are needed to determine which specialties are more likely to be subjected to WPV. Fourth, without data, we could not determine if violence prevalence differed by training level (residents vs. attending). Finally, future research should assess if their institutions have effective reporting procedures. Based on these findings, we need further well-designed research that considers all of the above to accurately assess WPV against HCWs.

Conclusion

The prevalence of medical staff who experienced some form of WPV, whether it was verbal (48%), physical (9%), or emotional (26%), was increased. Increases in WPV (40–47%), physical violence (12–23%), and verbal violence (45–58%) were observed from the middle to the end of the pandemic. The rate of physical violence was more than two times as high among nurses as it was among physicians (13% vs. 5%), whereas the rates of WPV and verbal violence were comparable. There was no correlation found between gender, profession, or COVID-19 timing with the probability of WPV, physical, or verbal violence. There was an increased risk of assault against COVID-19 healthcare workers. Our study limited the subgrouping of data according to the number of violent events, the specialty sparing, and the level of training. In general, this study provides substantial support for the decrease of violence as well as the protection of medical facilities and the safety of the workers. This could be accomplished by adopting prohibitive regulations by the government and raising awareness among the public.