Perception of Workplace Violence in the Emergency Department.

BACKGROUND: Workplace violence (WPV) is a serious issue worldwide. Violence against emergency department (ED) staffs causes significant physical and mental distress which affects work productivity and patient care.
OBJECTIVE: We seek amount and type of WPV perceived by the emergency physicians and nurses, their reporting agencies, and impact of violence on them.
METHODS: It was a cross-sectional study conducted at a tertiary health care center. Data were collected based on 24-item questionnaire between January and December 2017. Descriptive statistics was used to describe characteristics of participants and exposure to violence. Chi-square and Fisher's exact tests were used for bivariate analysis while logistic regression analysis was to analyze the impact of violence with participant characteristics. P < 0.05 was used to judge the clinical significance.
RESULTS: Two hundred and thirty-five participants (123 doctors and 112 nurses) completed the survey. About 67% of the participants (158/235) reported verbal abuse (VA), physical assault (PA) was reported by 17% (40/235), while confrontation was reported by 11% (27/235). Family members were the main perpetrator for VA (75%) and PA (35%). Regarding reporting, the violent incidences were mostly reported to ED security and ED faculty. Individuals with comparatively less age group, less experience, and male gender were more exposed to abuse both VA and PA at P < 0.05. Nurses and junior residents reported more abuse than senior residents (P < 0.05). Majority of the participants had reported lack of job satisfaction due to Verbal abuse (P = 0.01).
CONCLUSION: WPV is common in ED of the current setting. It results in significant physiological and psychological effects on health-care providers.

INTRODUCTION

Violence at work is a serious and concerning issue worldwide. According to the Bureau of Labor Statistics, 15% of all work-related fatalities are due to assaults and violent acts.[1] Exact figures on workplace violence (WPV) are difficult to quantify as these issues are often underreported.[12]

Epidemiological studies have identified emergency departments (EDs) as high-risk settings for violence against health-care workers. The 24-h unrestricted “open-door” policy, volume of patients, acuity of illness, and political focus all make ED staff vulnerable for violence.[3]

The National Institute for Occupational Safety and Health defines WPV ranging from offensive or threatening language to homicide.[4] It is not possible to document the violence in a proper way, primarily because of lack of any standard measurement instrument and institutional reporting policies, and second, because of the stigma attached to being a victim of violence. However, an overall evaluation of the findings from all studies clearly demonstrates that the level of violence in EDs is alarming.[5]

Emergency service providers have reported substantial and significant violent behaviors. To date, few studies have compared perceptions of WPV, and little data exist regarding the same in the health-care sector in India. In view of this, the investigators have created a survey to assess perceptions of WPV in the ED of a tertiary care hospital. The purpose of the study was to identify the incidence and characteristics of WPV and its impact among health-care professionals (doctors and nurses) working in ED.

METHODS

Study design

The cross-sectional survey approach was carried out among health-care professionals working in ED in a tertiary care hospital in Delhi from January to December 2017. The ED has screening room from where the patients get shifted to medical and surgical emergency units. The average number of patients visits to emergency is around 600–800 patients per day (according to hospital census 2016). The average nurse-to-patient ratio is 1:4 for red triage area and 1:20 for yellow and green triage areas. The average doctor-to-patient ratio is 1:15.

Selection of participants

The health-care professionals (postgraduate students, junior residents, senior residents, and nursing staff) working in ED dealing with direct patient care were included in the study, while the participants who were on leave or not willing to participate in the study were excluded. For the present study, total enumeration sampling technique was included. The total sample size calculated was 262, but only 235 were included due to nonavailability and not showing willingness to fill questionnaire for rest 27 individuals [Figure 1]. Among these 235, 123 were doctors (postgraduate students, junior residents, and senior residents) and 112 were nurses.

Figure 1

Consort diagram for sample size

Method of measurement

The survey carried out over 12 months assessed the WPV using a 24-item questionnaire which had items related to demographics, types, and numbers of violent acts encountered in ED and its impact on staff. The three types of violence were defined on the survey: verbal abuse (VA), physical assault (PA), and confrontations. A verbal threat was an oral communication by the patient or family member that was directly and specifically menacing to one's well-being. PA was an incident in which there was physical contact of an unwanted nature that resulted in harm or was perceived as threatening (i.e., being punched, kicked, bitten, pushed, and grabbed). A confrontation outside the immediate patient encounter was any unpleasant threatening interaction with the patient or other person representing the patient after the time one was actually providing medical care (i.e., telephone call, letter, or physical confrontation). The adjusted questionnaire was reviewed by five experts from the field of emergency medicine for face validity, content validity, and sensitivity of the items. The test–retest reliability for this questionnaire was established by determining stability over time, and the percentage of agreement was 94%, indicating good consistency.

Data collection and processing

Before data collection, ethical permission was obtained from the Institute Ethical Committee having reference number IEC/139/2/2016. Participants signed consent forms and were assigned individual identification numbers. The participants were told to fill the questionnaire at the end of their shift. The data collected were entered in MS Excel and analyzed using Strata 12. Descriptive statistics was used to describe the characteristics of participants and exposure to the type of violence. All results are presented as counts and valid percentages. In bivariate analysis, Chi-square test was used to find the association of WPV with demographic characteristics and with impact. Logistic regression analysis was used to analyze odds ratio for impact on demographics. A level of 0.05 was used to judge statistical significance.

RESULTS

Out of 235 participants, 123 were doctors and 112 were nurses working as full-time employees in emergency medicine department. Among the total participants, 156 were male and 79 were female.

Description of violent events and its reporting and impact

The study participants (n = 235) reported 225 violent episodes during the study period in the form of VA, physical abuse, and confrontation with varied frequency, as shown in Table 1.

Table 1

Frequency of workplace violence in emergency department (n=235)

Frequency	Verbal abuse, f (%)	Physical abuse, f (%)	Confrontation, f (%)
< a week	28 (18)	0	0
< a month	116 (73)	29 (72)	10 (37)
> a month	13 (9)	11 (28)	17 (63)
Total	158	40	27

Among total of 235 participants, 67% of the participants (n = 158) experienced VA, 17% (n = 40) physical abuse, and 11% (n = 27) experienced confrontation with violent episodes.

However, in context of its frequency, VA was experienced more frequently, i.e., < a week by 18% of participants and < a month period by 73% of participants, followed by physical abuse by 72% of participants in the same time frame, whereas frequency of confrontation was reported mainly in > a month period by 63% of participants. Regarding the perpetrators for violent events, as shown in Figure 2, the patient and patient's family members were mainly involved in violent events. Of these, patient's family members were mainly reported as perpetrators for VA by 75% of participants and for physical abuse by 35% of participants. The patient himself/herself was reported as perpetrator for VA by 27% of participants and for physical abuse by 5% of participants. ED staff member and other hospital staff were also claimed to be verbal abusers as reported by 19% and 12% of participants, respectively.

Figure 2

The common perpetrators for violent events in emergency department

Regarding reporting of WPV, the most common reporting channel was ED security (32% for VA and 15% for physical abuse) and ED faculty (26% for VA, 14% for physical abuse, and 10% for confrontation), as shown in Table 2.

Table 2

Reporting of workplace violence in emergency department (n=235)

Reporting bodies	Verbal abuse, f (%)	Physical abuse, f (%)	Confrontation, f (%)
Hospital police	8 (3)	13 (5)	5 (2)
ED faculty	62 (26)	34 (14)	23 (10)
Hospital administration	28 (12)	27 (11)	3 (1)
ED security	76 (32)	37 (15)	1 (0.5)

ED: Emergency department

The impact of WPV was observed in terms of effect on physiological and psychological health, as shown in Figure 3. The most common impact reported by majority was decreased job satisfaction in 54% of participants (n = 129), followed by feeling scared in 37% of participants (n = 87), loss of sleep in 29% of participants (n = 67), and least impact as missing the work due to violent incidences by merely 6% of participants (n = 15).

Figure 3

Impact of workplace violence on the study participants

The association of demographic variables was analyzed with VA, physical abuse, and confrontation, as shown in Table 3.

Table 3

Association of workplace violence with demographic variables (n=235)

Demographics variables	n	Verbal abuse, f (%)	P	Physical abuse, f (%)	P	Confrontation, f (%)	P
Age (years)
20-30	173	117 (74)	0.95	24 (63)	0.005*	12 (44)	0.451
30-40	51	34 (21)		9 (24)		10 (37)
>40	11	7 (5)		5 (13)		5 (19)
Gender
Male	156	100 (63)	0.151	26 (68)	0.771	24 (88)	0.089
Female	79	58 (37)		12 (32)		3 (12)#
Profession
Postgraduate students	18	10 (6)	0.182	2 (5)#	0.46	3 (12)#	0.215
Junior residents	84	52 (33)		10 (26)		10 (37)
Senior residents	21	13 (8)		4 (10)#		5 (19)
Nursing staff	112	83 (53)		22 (59)		9 (32)
Experience (years)
<1	70	82 (52)	0.559	20 (52)	0.003*	15 (56)	0.132
1-3	126	47 (30)		13 (34)		10 (37)
>3	39	29 (18)		5 (14)		2 (7)#

#Fisher’s exact test (if frequency<5)/Chi-square test, P<0.05 significant

Participants with more age and higher experience had less episodes of WPV with statistical significance (P < 0.05) but only with physical abuse. Males had experienced more violent episodes than females, but it was not found statistically significant. Similarly, nurses and junior residents had experienced more violent episodes than postgraduate students and senior residents, but it was also not found statistically significant with P = 0.05. The odds ratio was calculated by logistic regression, as shown in Table 4.

Table 4

Odds ratio for association of physical abuse with selected variables

Demographics variables	OR (unadjusted/crude)	CI	P	Adjusted OR	CI	P
Age (years)
20-30 versus >40	3.17	1.46-7.28	0.001*	1.57	0.34-7.14	0.005*
30-40 versus >40	1.33	0.57-3.47	0.50	1.21	0.22-1.72	0.36
Experience (years)
1-3 versus <1	0.27	0.12-1.07	0.01*	0.25	0.11-0.71	0.008*
>3 versus <1	0.61	0.13-1.88	0.394	0.37	0.11-1.19	0.09

OR: Odds ratio, CI: Confidence interval, Unadjusted Odds ratio For Age (20-30 years versus >40 years) P=0.001*, Adjusted Odds ratio For Age (20-30 years versus >40 years) P=0.005*

The adjusted odds ratio with 95% confidence interval was estimated by multiple logistic regression after controlling for age and experience. As described in table 4 odds of reporting physical abuse was higher in subjects with age group 20-30 years (odds ratio 1.57; Confidence Interval 0.34-7.14) than subjects with 30-40 years age group (odds ratio 1.21; Confidence Interval 0.22-1.72).

The odds ratio (adjusted value) of having physical abuse in participants with experience 1–3 years versus <1 year is 0.25 (P = 0.008) and >3 years versus <1 year is 0.37 (0.09), which means that participants with 1–3 years' experience are 73% less risk than <1-year experience to experience PA.

The association of WPV was also observed in terms of impact on participants, as shown in Table 5.

Table 5

The association of workplace violence with impact on participants

Workplace violence	Decreased job satisfaction, f (%)	P	Feeling scared, f (%)	P	Loss of sleep, f (%)	P	Missing the work, f (%)	P
Verbal abuse	94 (73)	0.05*	62 (71)	0.38	49 (73)	0.28	12 (80)	0.27
Physical abuse	21 (16)	0.99	14 (16)	0.99	10 (15)	0.84	2 (13)#	0.99
Confrontation	14 (11)	0.67	11 (13)	0.99	8 (12)	0.75	1 (7)#	0.98

#Fisher’s exact test (if frequency<5)/Chi-square test, P<0.05 significant, (Verbal abuse with decreased job satisfaction) P=0.05*

The findings suggested that VA is the most commonly reported factor for the impact on participants. Majority of the participants, i.e., 129 out of 235 had reported decreased job satisfaction which was mainly contributed by VA (73% of participants) with significant P = 0.01. Similarly, among total 235 participants, 87 participants felt scared, 67 participants had loss of sleep, and 15 had to miss the work mainly because of VA followed by physical abuse and confrontation but with no significant association at P = 0.05.

The association of demographic variables was observed with impact of WPV, as shown in Table 6. It states that age, profession, and experience were significantly associated with low job satisfaction, fear, and loss of sleep with P < 0.05. All types of impacts were less reported by participants with age, experience, and profession on higher side.

Table 6

Association of impact of workplace violence with demographic variable

Demographics variables	Low job satisfaction, f (%)	P	Feeling, scared, f (%)	P	Loss of sleep, f (%)	P	Missing work, f (%)	P
Age (years)
20-30	112 (87)	0.01*	72 (83)	0.004*	61 (91)	0.001*	12 (80)	0.69
30-40	15 (12)		15 (17)		6 (9)		2 (13)
>40	2 (1)#		0		0		1 (7)#
Gender
Male	86 (67)	0.99	58 (66)	0.99	43 (64)	0.650	12 (80)	0.39
Female	13 (33)		29 (34)		24 (36)		3 (20)#
Profession
Postgraduate students	12 (11)	0.001*	7 (8)	0.001*	7 (10)	0.003*	3 (20)	0.24
Junior residents	59 (45)		47 (55)		34 (51)		6 (40)
Senior residents	2 (1)#		1 (1)#		2 (3)#		1 (7)#
Nursing staff	56 (43)		32 (36)		24 (36)		5 (33)
Experience (years)
<1	65 (52)	0.02*	60 (69)	0.001*	58 (86)	0.001*	11 (74)	0.56
1-3	49 (30)		23 (26)		6 (9)		2 (13)#
>3	15 (18)		4 (15)		3 (5)#		2 (13)#

#Fisher’s exact test (if frequency<5)/Chi-square test, P<0.05 significant, (Age with feeling scared as impact) P=0.004*

DISCUSSION

Violence in the workplace is a common phenomenon. Numerous studies show that health-care professionals are victims of both verbal and physical violence. This is particularly true in high volume and trauma intensive EDs.[456] In studies of physician experiences in developed countries, violence was found to be part of the workplace environment.[4] In a study done in 1995 of 221 hospitals in North America, the ED was the most commonplace in the hospital for violent incidents.[7] In a systemic review based on 18 articles on WPV, 9 out of 18 studies (50%) reported that 53%–90% of medical personnel have experienced WPV. Physical or verbal violence was reported in 12 studies (67%), while PA was reported in 5 studies (28%).[8] As evident from the present survey findings, WPV was highly prevalent among the ED health-care staff, highlighting the seriousness of the issue. This study adds to the body of literature supporting the argument that health-care workers working in EDs are exposed to greater abuse in their workplaces, and WPV has detrimental effects on both health workers and the health-care system. The present study claimed high incident rate of 67% for VA, followed by 17% for physical abuse and then 11% for confrontation. A study done in ED in Vancouver showed 92% prevalence rate for PA and 97% for verbal threats which are on the higher side to the present study.[9] In another study from Turkey, 78.1% of the emergency physicians reported being subjected to violence out of which 65.9% had suffered on more than one occasion over the past 1 year.[10] In a similar kind of study done in ED of Morocco, 70% of health-care professionals had been exposed to WPV; out of those, 77% of the participants stated that they experienced VA or verbal threat, while 8.3% reported about PA.[11] The findings of this study are also consistent with the study done in emergency unit in Ismailia in 2017, which reported that patient's relatives were the main perpetrators of VA and physical abuse against emergency health-care workers.[12] The majority of the participants of this study experienced VA very frequently (< a week by 18% and < a month by 73% of participants) as compared to physical abuse. On the other hand, the confrontation after patient care was experienced quite infrequently (> a month period). In comparison, the findings of a similar study on frequency of WPV on emergency physicians reported that 32% of the respondents experienced at least one verbal threat per day, and 18% noted that weapons were displayed in a threatening manner at least once per month.[13] A similar kind of study was conducted at a tertiary health care centre in the USA to identify WPV among resident doctors which demonstrated that majority of the participants (96.6%) had VA or verbal threat, while 65.5% experienced physical violence in the ED and most of the violence was committed by the patient rather than the attendant or their relatives which was contrary to the findings of the present study.[14] No protective or harmful effects in terms of years of experience were seen in this study.[14]

The reporting of violent incidences was mainly to ED faculty and security staff in our study. On the contrary in a study, over half of the respondents (57%) said that reporting was useless and 14% did not report because they were afraid of negative consequences. The main impact reported by majority of the participants in this study was decreased job satisfaction (54% of participants) and fear (37% of participants). A systemic review based on 18 observational studies showed that 44% of the participants in ED had negative impacts of WPV in the form of threatened job safety, increased sick leave, decreased job satisfaction, poor decision-making, burnout, and decreased well-being. The studies also revealed that the emergency care providers had loss of joy in providing care and they used to spend less time with such abusive patients or their relatives.[815] Gordon et al. found that WPV has a direct relationship to stress, decreased work productivity, and quality of patient care. The present study findings reported a higher incidence of violence and its impacts among participants with less age group, less experience, male gender, nurses, and junior residents; however, a significant association (P < 0.05) was found only for physical abuse with age and experience. In a similar type of study, it was found that physicians below the age of 30 years with ED experience of 1–4 years had the greatest odds of experiencing violence. A statistically significant relationship was found between physicians' current position and experience in EDs and participant to violence. This observation was very similar to our study.[1015] A study conducted in Saudi Arabia in 2012 found that nurses were more likely to be exposed to violence (P < 0.001) than other hospital staff.[15] In association of impact with violence, VA had a significant association for decreased job satisfaction with P = 0.04 [Table 5], whereas another study quoted verbal and physical abuses as main factors for high turnover and increased absenteeism from emergency ward. Regarding reporting, the incidences were majorly reported by nurses in the present study which is contrary to a study which stated only 3 out of 72 nursing participants reported the incidents of physical abuse to a senior staff member or ward in-charge.[16]

Limitation of the study

It is a single-center study which was limited to only doctors and nurses. The responses were based on close-ended responses of the participants rather real-time observations. The cross-sectional nature of this survey means that no causal time-related effects drawn from it.

CONCLUSION

The findings of the study showed that there was a high prevalence of WPV in ED, with the majority of the incidents were in the form of VA/threats. Most of the violent incidences were committed by patient's family member. The reporting of WPV was very less as compared to actual incidences which seriously need to be addressed. Professionals with lesser age and experience are more vulnerable. In addition, nurses and male gender were at higher risk to be victim of such violent incidences. Decrease in job satisfaction was the most commonly reported impact because of WPV.

Implications and recommendations

Further research is needed to compare the incidence of physical and nonphysical types of WPV among the health-care workers working in public and private sectors. There is also a need of further research to find the factors responsible for such violence and its reporting to authorities. There is also a need to plan educational interventions or training to empower our health-care workers working in acute care units. All health-care workers should be encouraged to report every violent act promptly. Every health-care worker should be sensitized about WPV through their in-service education or induction programs. Measures to alleviate overcrowding, improved working condition, limited entry of attendants, increased security of ED, and zero tolerance to violence are the few suggested methods to prevent violence in ED. WPV has a lot of factors and complexity; hence, it could be effectively tackled with strategies based on “multidimensional” analysis of the operating ambiences and interventions.[17]

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.