5

Work 51 (2015) 5–18 DOI 10.3233/WOR-141893 IOS Press

Staff perspectives of violence in the emergency department: Appeals for consequences, collaboration, and consistency Paula Renker, Shellie A. Scribner∗ and Pam Huff Grant Medical Center, Columbus, OH, USA

Received 12 February 2013 Accepted 14 July 2013

Abstract. BACKGROUND: Violence committed by patients and their families and visitors against Emergency Department staff in the United States is common and detrimental to staff well being, morale, and care practices. Hospitals losses occur due to decreased staff retention, prestige, and patient and visitor satisfaction. OBJECTIVE: The purpose of the baseline survey reported here was to identify and describe staff experiences, concerns, and perceptions related to violence and abuse perpetrated by patients, family, and non-family visitors in a Level 1 emergency department. PARTICIPANTS: The survey sample was composed of 41 registered nurses and 10 paramedics. The majority of the participants (84%, n = 41) were female and worked full time (82%, n = 41) on the 7P-7A (49%, n = 25) shift. METHODS: The cross-sectional mixed-method descriptive design used a survey to measure violence experiences and interviews with key informants. Specific analytical methods included descriptive and inferential statistics and ethnography. RESULTS: The findings are summarized by a model that portrays 1) Contributing factors to the development of violence in the ED, 2) maladaptive reactions to workplace violence of Cynicism, Concern for focus on customer service, and Conflict, and 3) three themes that, depending on their presence or absence, serve as barriers or facilitators to violence: Consistency, Consequences and Collaboration. CONCLUSIONS: Interventions developed to minimize violence in the ED must focus on modifiable risk factors and address what is in the department’s control including staff education in recognizing escalating anxious or aggressive behavior, policy development and implementation, and environmental changes. Keywords: Nurse and paramedic patient interactions, mixed methods analysis, workplace violence interventions

1. Introduction Workplace violence is at the forefront of priorities for assessment and intervention among national and governmental organizations including The Joint Commission, the American College of Emergency Physi∗ Corresponding author: Shellie Scribner, Grant Medical Center, 111 South Grant Avenue, Columbus, OH 43215, USA. Tel.: +1 614 566 7067; Fax: +1 614 566 8247; E-mail: shellie.scribner@ ohiohealth.com.

cians, the Emergency Nurses Association, and the American Nurses Association [1–4]. The Occupational Safety and Health Administration defines workplace violence as “any act or threat of physical violence, harassment, intimidation, or other threatening disruptive behavior that occurs at the work site. It ranges from threats and verbal abuse to physical assaults and even homicide. It can affect and involve employees, clients, customers and visitors” [5]. Workplace violence against emergency department workers is common in the United States with several studies report-

c 2015 – IOS Press and the authors. All rights reserved 1051-9815/15/$35.00 

6

P. Renker et al. / Staff perspectives of violence in the emergency department

ing that physical violence against nurses can range from 42–67% over 6–12 months and as high as 72% over their lifetime [6–9]. Even with multiple studies reporting a high prevalence of violence, the Emergency Nurses Association proposes that workplace violence against nurses in the U.S. is vastly underreported [10]. Underreporting limits full awareness of the extent of violence against emergency nurses in the workplace. Kowalenko et al. [11] propose that experiencing physical violence and verbal abuse confounds the stress health care workers already experience in their day to day jobs in caring for victims of injury and violence. Nurses are subjected to both short and long term effects of violence including physical consequences related to lacerations, bruises, and musculoskeletal strain or damage and emotional sequelae of Post Traumatic Stress Disorder [PTSD], anger, fear, shame, and difficulty maintaining concentration [10,12,13]. Whether short or long term, the negative effects of violence influence job performance and nurses’ abilities to care for patients [13–17]. In addition to jeopardizing personal safety, workplace violence may also decrease productivity and job satisfaction and lead to early burnout [11,13,17,18]. Besides the debilitating physical and emotional effects for staff members, addressing workplace violence has a global impact on all aspects of the business of healthcare. Workplace violence is costly in terms of lost work days. The Bureau of Labor Statistics states that registered nurses account for 46% of all nonfatal assaults and violent acts related to workplace violence requiring days away from work [19]. Along with lost work days, departmental budgets are also affected by decreased productivity, decreased retention and increased turnover, medical and psychological care for employees, and workers’ compensation [12,17,18,20]. The hospital also bears costs related to decreased patient satisfaction, increased litigation, and changes in hospital reputation [10,13]. While several integrative and systematic reviews conclude that the prevalence of workplace violence is well established in the literature, they also find that there is minimal evidence of the development and efficacy of interventions [21,22]. Important research opportunities exist in the area related to understanding safety-related perspectives of all staff members so that interventions can be developed and tested. The purpose of this study was to identify and describe staff experiences, concerns, and perceptions related to violence and abuse perpetrated by patients, family, and non-family visitors. This study is a part of

a larger study to improve emergency department (ED) healthcare providers’ safety in the workplace. The results from this study will provide for the development of culture-specific interventions to enhance health care providers’ safety and job satisfaction. In addition, ED patients’ and visitors’ hospital experiences will be improved when receiving care in a safer environment.

2. Methods The research team was comprised of the Primary Investigator (PI) who conducted data entry and analysis and two ED registered nurse (RN) Clinical Educators who assisted with data collection and data interpretation. 2.1. Design A cross-sectional mixed-method descriptive design was used to address the research questions associated with this study. The study included a standardized survey with both quantitative and open-ended items to measure violence experiences and concerns as well as interviews with key informants to afford an in-depth understanding of administrative and staff perspectives. 2.2. Sample The study site is a nongovernment, not for profit, teaching adult hospital without a separate psychiatric department. It is also designated as a Level I Trauma Center by the American College of Surgeons. Inclusion criteria for the survey included all ED registered nurses, paramedics, unlicensed assistive personnel (UAP), and unit clerks employed in the hospital ED in the Midwest U.S. One hundred and forty-two staff members were eligible for the survey. In addition, several ED nurse administrators and ED staff nurses spoke with the primary investigator about their experiences, perceptions, and insights regarding the complex situations and nurse-patient relationships that are affected by workplace violence. As identified previously, nursing administrators were not included in the survey – it is not known if the staff nurses who spoke to the author had completed the survey due to the anonymity extended to all survey participants. While the discussions with ED personnel and administrators were not planned in the original research proposal, the PI asked each of the individuals if their comments could be used as part of the study. These self-selected volunteers became key-informants and provided an additional source of data for the analysis.

P. Renker et al. / Staff perspectives of violence in the emergency department

2.3. Research questions The following research questions were addressed in this study: 1) What are staff members’ experiences with physical violence and verbal abuse initiated by patients, patient’s family, and/or non-family visitors in terms of types and frequency? 2) What are staff perspectives of initiating factors or causes of workplace violence? 3) What are staff perspectives of security measures in place in their department? 4) What are staff perspectives of policies, interventions, and consequences related to workplace violence? 2.4. Measures The survey used in this study was developed by the Emergency Nurses Association (ENA) for use in the study “Violence Against Nurses Working in U.S. Emergency Departments” and used with the author’s permission [23]. The survey is a 39-item survey concerning the respondent’s personal experiences with physical violence and verbal abuse from patients and/or visitors in the ED, the policies and procedures of the participant’s hospital and ED for addressing workplace violence, and participant’s beliefs about adequacy of available security, the precipitating factors of violence from patients/visitors and barriers to reporting violence in the ED. The ENA established face validity for the survey through review by a panel of 15 content experts who were asked to complete the survey and comment on the relevance and clarity of the questions. A total of 3,465 emergency nurse participants completed Gacki-Smith’s et al. survey (Personal written correspondence Jessica Gacki-Smith, September 14, 2010). The ENA survey [23] includes questions in multiple formats including rating scales, multi-item check lists, multiple choice items, numerical rating scales in Visual Analogue Scale (VAS) format, yes/no, Likert, and open-ended questions. Demographic variables for this study included: age, gender, education, scheduling, and clinical experience in years. Two versions of the survey were used. Nurses completed the original survey. A second version of the survey was developed for non-nurse participants (UAP, paramedics, and unitclerks) by altering the wording from “nurse” to “staff” and adapting demographic questions related to education and certification to reflect appropriate categories

7

for non-nurses. While the original survey by GackiSmith et al. was web-administered, the survey used in this study was conducted in a paper and pencil format. All other items and formatting were unchanged. Keyinformants were asked to reflect on their perspectives of causes of violence, barriers to reporting, changes needed to address violence, and nurses’ response to violence. Table 1 relates how the items in the ENA survey and key informant responses are related to this study’s research questions. 2.5. Analysis 2.5.1. Mixed method analytic approach The purpose of a study and the associated hypotheses and/or research questions is to drive the paradigmatic approach to data analysis [24,25]. Because the purpose of our study encompassed identifying the prevalence of, and perceptions related to, violence experienced in the ED it was most advantageous to use both qualitative and quantitative analytic approaches in a mixed-method design. The survey used for the study was well suited to address the research questions and design because it included both fixed-choice and open-ended items. While one of the open-ended items (“What is the one most important policy, procedure, or environmental control your ED uses, or could use, to prevent workplace violence against RNs?”) was written in a manner that could have been analyzed by counting the frequency of responses, participants often included lengthy narratives related to their reasons for choosing their response. In addition, participants’ responses to the second open-ended item (“What other comments do you have, or issues you would like to addressed, concerning violence against RNs in the ED?”) were all narratives of experiences, perceptions, concerns, and needs. Other sources of qualitative data included narratives and comments added by participants adjacent to the fixed-choice survey items, and information spontaneously shared by ED staff and administrators during discussions with one of the study investigators (PR). Mixed method analyses were used in this study not only because of the need to analyze both qualitative and quantitative data (that could have been addressed in a multi-method design) but because of the method’s ability to address the need to incorporate or to integrate findings from both the deductive (quantitative) and inductive (qualitative) paradigms to enhance the findings’ dimensionality, scope, and utility [24,25]. The theoretical drive (or plan of mixed method analysis) for this study was simultaneous – or to collect

8

P. Renker et al. / Staff perspectives of violence in the emergency department Table 1 Survey and interview items informing the research questions Question RQ 1:

Items What are staff member’s experiences with physical violence and verbal abuse initiated by patients, patient’s families, and nonfamily visitors? 1. Rating scale (1-10) of the overall safety of personnel (as opposed to their personal safety). 2. Items related to identification (occurrence and frequency) of multiple forms of physical violence and verbal abuse. 3. Items related to past and present levels of violence in ED.

RQ2:

What are staff perspectives of precipitating or initiating factors or causes of workplace violence? 1. Twenty-nine item “all that apply” checklist of possible precipitating factors to workplace violence plus one open-ended item. 2. Two items related to staff perspectives of violence being an expected occurrence in ED care.

RQ3:

What are staff perspectives of security measures in place in their department? 1. Rating scale of the adequacy (effectiveness) of the amount of security (1–10). 2. Rating scale (1–10) of the overall safety of personnel (as opposed to their personal safety). 3. Likert-rating evaluation related to the presence and efficiency of 23 environmental controls providing ED security plus one open-ended item.

RQ 4:

What are staff perspectives of policies, interventions, and consequences related to workplace violence? 1. Items related to staff education for prevention or diffusion of violence in the ED i. Items “Rate how well the (educational) courses met your needs as an emergency nurse” and “Explain why this/these courses did not completely meet your needs”. ii. Items related to frequency of discussion in staff meetings about violence. 2. Items related to policies that address violence i. Existence of Policies. ii. Guidance related to implementing policies. iii. Barriers related to reporting violence (16 item checklist plus one open-ended item). 3. Items related to administrative (hospital and departmental) commitment to eliminate violence. 4. Hospital’s response or recommendation to a staff member who reports that a violence incident was directed at him/her. (10 item “all that apply” checklist plus open-ended item.) 5. Open-ended: “What is the one most important policy, procedure, or environmental control your ED uses or could use, to prevent workplace violence?” 6. Open-ended “What other comments do you have, or issues you would like to see addressed, concerning workplace violence in the ED”. 7. Comments included on surveys and interviews about work place violence including, staff responses, associated negative outcomes, personal experiences, and consequences of violence for staff and patients

data on the phenomena from both quantitative and qualitative sources at the same time with the qualitative approach being the initial port of inquiry and the quantitative findings supporting or enhancing the qualitative findings. Morse depicts this type of analysis as Qual + Quan [25]. Initial separate analyses of quantitative and qualitative data allow the researcher the ability to identify patterns and relationships between the findings from each approach which are then subsequently merged in the research findings. Morse graphically portrayed the Qual + Quan research process used in this study as containing two pathways portraying how the two analytic processes are incorporated into the research process and findings (Fig. 1) [25]. 2.5.2. Left pathway: Qualitative analysis Cresswell’s [26] and Walcott’s [27] approaches to analyzing ethnographic data were used in this study. Ethnographic analyses reflect participants’ actions, be-

liefs, and experiences in the context of a specific culture, such as the culture of the ED. The data used in this study emerged both from the written survey and discussions with key informants. Creswell [26] describes six key aspects of ethnographic analyses that were used in this study. The first four steps, data management, reading and memoing, describing, and classifying were undertaken as qualitative analysis in the left analytical pathway and are described below. Data management: Responses from the open-ended items were typed and recorded into separate files for analysis [24]. Individual items written by participants next to the quantitative items in the survey were similarly entered into a separate file organized by survey item. Field notes were taken after interviews with the key informants and later transcribed by the interviewer. Reading, memoing: After the data were assembled for review, each file was read through several times. After the researcher became familiar with the content,

P. Renker et al. / Staff perspectives of violence in the emergency department

9

Research Question Left Pathway Right Pathway Theoretical drive QUAL (Inductive)

Core Component of Project Select core (primary) method

Enhancing description, understanding or explanation or testing conjectures by including a QUAN (Deductive) analysis

Supplemental Component of Project Select appropriate supplementary strategy and follow procedures to maintain validity Simultaneous /Sequential supplementary strategies for QUAL +QUAN

Collect core (primary) data Collect supplementary data

Analyze data

Research Findings

Analyze supplementary findings with those of core component

Integrate supplementary findings with those of core component

Inform the research question

c 2005 [25]. Fig. 1. Overview of the process of Mixed-method design Adapted with permission of the author, Janice Morse 

codes or comments were made in margins about tentative responses or ideas about the meaning of statements. Describing: Creswell [26] relates the next step of analysis as describing the social setting, participants, and events. Walcott [27] proposes that data need to be displayed in varied formats and organizational approaches. The various data files from the different sources of qualitative data for this study were organized in a manner to enable the researcher to compare and contrast codes among and between the files. The

list of the initial comments and descriptors were then compiled into a separate data file. Classifying: The newly developed list of codes were then scrutinized to identify similarities or patterns. The comments were organized into another file into groups that reflected their characteristics. Codes or terminological references were developed that encompassed the meanings associated with each group of comments. New (raw) data files were then printed and the researcher used the newly developed codes as a template to evaluate their fit with the raw data. The files

10

P. Renker et al. / Staff perspectives of violence in the emergency department

were scrutinized for data that did not fit into one of the codes and the unassigned data were then examined to determine if one or more new codes needed to be developed to fully address all of the data. The process continued until the maximum amount of data were reflected by one of the codes. The resulting list of codes were then separated from the raw data and the coding process began a second time to reduce the total number of codes to a manageable amount (5–6) of themes that parsimoniously, but sufficiently, described the data [26,27]. The newly developed list of themes was then displayed graphically and the process was reversed as specific pieces of raw data were placed on a chart beneath the codes [23]. This spiral approach to qualitative analysis facilitates the process of fully describing the data through repeated immersion into the data and codes [26]. 2.5.3. Right pathway: Quantitative analysis Quantitative data from fixed-response survey items were entered into and analyzed using Statistical Package for the Social Sciences [SPSS]17 [28]. Descriptive statistics were used to describe item responses and demographics. Chi-Square analyses and independent ttests were used to compare and contrast sample demographics with designated survey items and to compare and contrast responses between ancillary staff and registered nurses. Two groups of quantitative data analyses emerged with the first group addressing specific concepts related to the first three research questions and the second group supporting the general purpose of the study (to describe health care providers experiences with, and perceptions of, violence in the ED). Data related to the thematic analyses were then transformed and merged with the qualitative data core analyses [24]. The approach of quantitative analyses being used in two different manners as described above is not uncommon with mixed method designs [23]. 2.5.4. Integration of thematic analyses from the left pathway and quantitative analyses from the right pathway Creswell conceptualizes two different steps that foster the analysis of ethnographic data in mixed method analyses: 1) interpreting and 2) representing and visualizing [26]. The interpreting phase of analysis is where the researcher strives to make sense of the findings (in mixed method the analyses from the two different pathways) [24,29]. The process used in identifying codes and themes is repeated with the transformed quantitative data to supplement and enrich the

core qualitative analysis [23,24,26]. The last step in the analysis process encompasses representing and visualizing the data. Tables, graphs, pictures, and drawings were used to assist the researcher to develop and finalize the relationships between the various themes and to demonstrate how the themes work together to address the purpose of the study. A final model or representation of the study findings was created. The final report is represented by detailed (thick) description of the data that supports and explains each of the themes. Quotes from participants are used as well as the researcher’s interpretation of underlying factors or situations that might have formulated or influenced the participants’ sentiments. The final report also includes a detailed description of the model or summative account that depicts the relationships between the themes [24–26]. It is important to note that the qualitative process of considering the researcher’s interpretation of the study findings as results lies in opposition to the traditional (quantitative) approach of reporting data summaries and statistical analyses in the result section and interpretation in the discussion section. 2.6. Human subject information The proposal and Institutional Review Board (IRB) protocol were submitted to the hospital’s IRB and were deemed to be exempt. The researchers, however, included a statement about the rights of the nurses as research subjects at the beginning of the written survey including the provision of anonymity, that results would be published in a manner to protect the identification of any one individual, and that their participation was voluntary. The use of additional data streams in qualitative research as with the spontaneous interviews in this study is a unique aspect of qualitative designs. The individuals contributing information to the study volunteered for their comments to be included in the study and were offered confidentiality by the researcher. 2.7. Data collection The research study was described to potential participants during staff meetings by one of the staff members of the study team. Surveys were placed in the mailboxes of each registered nurse, UAP, paramedic, and unit clerk. A locked wooden box with a surveysized opening was placed in the break room used by all personnel to deposit completed surveys. One round of surveys was conducted over a 4 week time period.

P. Renker et al. / Staff perspectives of violence in the emergency department

11

Table 2 Health care providers experiences with physical violence ranked by total experiences with violence and identity of perpetrator(s) (n = 52) Physical violent act Hit Pinched Spit on Kicked Scratched Pushed/shoved Hit by thrown objects Voided on Hair Pulled Biting Sexually assaulted Shot Stabbed

Total experiences with violence 72.6% 68.7% 66% 62.8% 60.9% 49% 39.2% 38.8% 30.6% 29.4% 6% 0% 0%

By patient 66.7% 66.7% 54% 60.8% 56.9% 31.4% 33.3% 38.8% 30.6% 25.5% 2%

Of the 142 surveys that were distributed to staff members via their unit mailboxes, 52 were returned with complete data and available for analysis with an overall return rate of 36.6%. The distribution of occupations of those returning the surveys was skewed however, with 41/101 RNs returning the surveys, 10/27 paramedics, and no UAP or unit clerks (one person submitted data without designating their occupation). The return rate of 40% for RNS and 37% for the paramedics is a more accurate reflection of their participation in the study, and the lack of participation from the UAP and unit clerks restricts any generalization of the study findings to their experiences.

3. Results 3.1. Sample description The majority of the participants (84%, n = 41) were female and worked full time (82%, n = 41). Mean years of employment at the hospital were 5.67 (± 7.23) and 4.25 (± 4.20) years for nurses and paramedics, respectively. In addition, nurses reported that their overall work experiences (including that in other hospitals) as a mean of 8.67 (± 8.25) years. Almost half of the sample (49%, n = 25) worked 7P-7A, 32.7% (n = 16) worked one of the day shifts, and the remaining worked the bridge shift (11A-11P). The sample characteristics represented the demographic characteristics of the ED staff. Because there were no significant differences between the responses of the nurses and paramedics to the quantitative survey items (using chi-square and independent t-tests) their data were combined for the remainder of the analyses.

By family

By patient and family 3.9%

By visitor

By all 2%

2% 10%

2% 2%

2% 11.8% 2%

2%

2% 3.8% 3.9%

2%

2%

3.9%

Has never happened 27.4% 31.3% 34% 37.3% 39.1% 51% 60.8% 61.2% 69.4% 70.6% 94% 100% 100%

3.2. Research questions The first research question asked staff members to identify the types and frequency of any physical violence and/or verbal abuse initiated by patients, patient’s family, and non-family visitors. In all, 96.1% (n = 50) of the participants reported experiencing physical violence with 39.2% (n = 20) reporting that they experienced physical violence at least weekly. The total increased to 68.6% (n = 35) when the time period was extended to monthly. All of the participants reported experiencing verbal abuse with 52% (n = 27) indicating that they experienced the assaults on every shift they worked. Two-thirds (n = 34) of the participants stated that they felt that violence had increased in the last year. Table 2 describes participants’ responses to the physical violence assessment items. The most common acts of physical violence included being hit, pinched, spit upon, kicked, and scratched (> 60%), while the lowest percentage (40%) reported being either voided on or kicked. Three participants in the study included additional types of injuries related to violence to the pre-established checklist. They described being threatened with an ice pick, being cornered in a room and not allowed to leave, and having a dislocated shoulder. The only type of violence perpetrated solely by the patient was being voided on. An even larger portion of our participants identified that they experienced verbal abuse with more than three out of four participants reporting they had been: 1) yelled or shouted at (100%, n = 51); 2) sworn or cursed at (100%, n = 51); 3) been called names (94.1%, n = 48); 4) threatened with physical harm (82.4%, n = 42); 5) harassed with sexual language and innuendo (76.6%, n = 39); and 6) intimidated (76.6%, n = 39). When asked about the frequency of verbal

12

P. Renker et al. / Staff perspectives of violence in the emergency department Table 3 Staff ratings of factors precipitating violence in the emergency department ranked in percentage of endorsement Hospital environment External environment – Location of the facility (82.4%) – Local violent crime rates high (56.9%) Internal environment – Crowding/high patient volume (76.5%) – Prolonged wait times (68.6%) – Holding/boarding patients (56.9%) Policies and attitudes – No smoking policy affecting patients and families waiting in ED (49%) – Staff removing personal items from patients (35.3%) Patient and family characteristics – Drug seeking behaviors (96.1%) – Patient/visitors under the influence of alcohol (94.1%) – Patient/visitors under the influence of illicit drugs (90.2%) – Psychiatric patients being cared for in the ED (84.3%) – Dementia/Alzheimer’s patients being cared for in the ED (48%) Staff characteristics of perceptions of staff – Poorly enforced visitor policy (64.7%) – Misconceptions of patients and visitors of staff behavior (56.9%) – Patient/visitors perspectives that staff is uncaring (45.1%) – Perceived prejudice from staff (41.2%)

abuse experiences, 52% (n = 27) reported that it occurred on each shift that they worked and the prevalence increased to 92% (n = 46) when weekly and monthly data were added. Similar to the perpetrators of physical violence, the patient was the perpetrator most often reported as the instigator of verbal abuse, although many of the acts were attributed to family, visitors, or all three. The second research question related to staff perspectives of initiating factors or causes of violence. The survey item presented a list of 30 alphabetized potential factors for precipitating violence with participants asked to identify all that they felt applied. The factors measured quantitatively generally clustered in one of four general areas: hospital environment (internal and external); policies and attitudes; patient and family characteristics; and staff actions and characteristics. Table 3 describes staff’s quantitative responses to precipitating factors of violence in their ED that had greater than 35% agreement. Items rated with less than 35% agreement included cultural barriers, shortage of quiet and seclusion rooms, language and literacy issues, lack of ED RNs and physicians, lack of patient privacy, not keeping patients informed, grieving family and visitors, and receiving cash payments. Several items contributed to the third research question regarding staff’s perspectives of available security procedures. Participants’ responses to the items rating the effectiveness of overall security and their perceptions of overall staff security are displayed in Table 4. Security measures were also evaluated using a rating

scale of 1 (“not at all efficient”) – 4 (“very efficient”). Each item could also be rated as “not used” or “not sure if used” in the ED. Only one of the security strategies, restraints, received a majority of participants rating “very efficient” (57%, n = 29). Perhaps one of the more perplexing findings in this section of the analysis was that several of the controls rated highest as “very efficient” were ranked by more staff members as “not sure if used” or “not used in ED”. Table 5 describes the relationships between the categories of highest ranking strategies for “very efficient”, “not at all efficient”, and “not used” or “not sure if used in ED”. Mixed–method analysis was used to address the final research question “What are staff perspectives of policies, interventions, and consequences related to workplace violence?” Seven themes emerged from the analysis of the responses to the open-ended questions, the participant-initiated comments and stories written on the survey, and the field notes from the key informant interviews. Participants’ words and stories help to provide a deeper understanding of each of the themes as described below. Theme 1: Contributing Factors Numerous participant responses combined to portray the Contributing Factors that participants attributed to violence in their workplace. Several commented on aspects of the physical environment that increased staff jeopardy such as multiple, unguarded entries into the unit, fear of being alone in an isolated area of the ED and a lack of bulletproof glass at the main entrance to protect them from gunfire. Multiple

P. Renker et al. / Staff perspectives of violence in the emergency department

13

Table 4 Staff ratings of overall effectiveness of hospital security and level of staff safety Item Rate the type of security your hospital has. . . Rate the overall level of staff safety from workplace violence in our ED

Scale 1 (Not at all effective) to 10 (Extremely effective) 1 (Not safe at all) to 10 (Extremely safe)

Results M = 5.64 (± 2.110, range 2–10, n = 47) M = 4.90 (± 2.013, range 1–9)

Table 5 Staff perspectives of efficiency and use of security items in the ED in terms of percentage of endorsement (n =51) Very efficient Restraints (57%) Security cameras (39.2%) Bullet-proof vests (35.3%) Panic button/silent alarm (33.3%) Pepper spray or mace (28%) Gun (27.1%)

Not at all efficient Locked/coded ED entries (23.5%) Personal belonging search (20%) Metal detectors (20%) Security signage (16%)

barriers to reporting violence increased participants’ frustration and included perceived lack of hospital administrative and managerial support (“nothing will be done”), no evidence of physical injury, an ambiguous reporting policy, and an “inconvenient and lengthy reporting procedure” to use when reporting. Several participants shared that they were frustrated because the only time they could complete violence reports was after work hours. Staff ratings of their perceived level of safety and efficiency of security measures identified above demonstrate their concerns for their safety as a contributing factor to violence in the ED. It is possible that participants’ lack of knowledge about existing security strategies may have contributed to their concerns about their safety. Another interesting insight relates to the factors that were perceived as most efficient were primarily implemented by security or ordered by physicians. None of the participants identified that they wanted to wear bullet proof vests, carry guns, or discharge mace. However, two comments by nurses about violence prevention educational programs helped to put these findings in perspective “I don’t want to learn gun safety and how to disarm someone. I’d probably be the one hurt in that case. We need a security officer to handle these cases so we can concentrate on caring for our patients” and “. . . it puts the entire burden on the nurse who is never the less expected to achieve pain relief; make patient[s] perfectly comfortable, etc.” In addition to concerns related to the physical environment, barriers to reporting, and diminished perceptions of security efficiency and their collective safety, the majority of the nurses perceived that the level of violence in their ED was associated with being in an urban area with

Not sure if used or not used Batons (62.7%) Safe for cash payments (60%) Mirrors (44.3%) Bulletproof vests (43.1%) Guns (40%) Pepper spray or mace (30%) Security signage (24%)

drug-seeking patients and family members and visitors under the influence of alcohol or drugs. Themes 2–4: Concerns for Focus on Customer Service, Conflict, and Cynicism. Three themes emerged from various behavioral reactions associated with the violence milieu that were identified by participants. Maladaptive behaviors and responses associated with these themes serve to perpetuate the milieu of violence. Comments related to Concerns for Focus on Customer Service may not be unique to the hospital setting. In this analysis, participants focused on the difficulty of meeting patient’s perceived needs when they differ from the plan of care. Their stories and concerns were often associated with passion, anger, disgust, and pessimism and included statements such as: “Patient satisfaction is more important than staff feeling safe and non-violated”; “The docs are so afraid that their [patient satisfaction] scores will be affected that they bend over backwards to meet every demand”; and “Patients are rewarded with footies, food. . . for bad behavior. This tells the patient that they are right and the behavior is repeated”. One participant summed other participants’ comments by saying “Patients are continually being referred to as “customers” – they are not customers, they are patients. As a customer in a restaurant or store I can order what I want and rightfully be upset if I do not receive what I ordered. As a patient you don’t order treatments, the hospital staff are the educated ones with their licenses “on the line” caring for the patient. Too many times I am cursed at, yelled at, berated for the fact that a patient feels they are not getting what they feel they should get i.e. pain med-

14

P. Renker et al. / Staff perspectives of violence in the emergency department

ication, narcotics, food, drinks, CT [computerized tomography], x-rays. I have even had family members of patients call in and yell (at me) based on what the patient has told them.” Another participant shared “It is not customer service to let patients and family be abusive toward staffit leads to poor, inconsistent care for the patient, and disturbs the other patients.” These perspectives of customer service may contribute to the angst associated with the themes of Conflict and Cynicism. Comments associated with the Conflict theme revealed uneasy work relationships with physicians and other colleagues. One paramedic participant wrote “I feel everyone but the Patient Support Associates [PSAs] [Unlicensed Assistive Personnel] and paramedics have the ability to have a violent or abusive patient thrown out”. Other participants shared “I believe a majority of physicians prescribe narcotics more frequently than normal. When a patient doesn’t get them (narcotics) they become violent” and “We condone bad behavior by giving meds (pain) to verbally and physically abusive people. We give them the satisfaction that this abusive behavior will get them what they want”. Another participant shared that “It is often more difficult to reduce tension/issues with trauma or admitted patients due to physician communication/or lack of.” Security officers were also the topic of many comments related to conflicts in the ED. “I have also seen some security officers not check purses/belongings if there is a confrontation. They also let crowds of visitors back despite repeated reminders.” Lastly, “To sum it up there is really poor visitor control and monitoring as well as screening for weapons. People can access us too easily and very easily pass thru security with a weapon.” Cynicism is not only a maladaptive reaction associated with the violent milieu in the ED, but is also seen by some as a contributing factor to the development of violent interactions. One participant queried “How do we deal with the cynicism that inevitably results in the RN?” Other participants raised concerns about staff “stereotyping” possibly resulting in missing critical patient care needs or in limiting patient interactions due to fear of getting hurt. Cynicism was also reflected in the number of nurses who responded that violence is “part of the job”. Others shared that “Policies and procedures don’t change even if reported, no one cares.” One participant poignantly shared “I don’t believe that any policy will change until someone is seriously injured or hurt. “

Themes 5–7: Consequences, Consistency, and Collaboration Participants also shared many perspectives of ways to address violence concerns and to enhance the safety of the workplace. Their perspectives were presented dialectically in that some participants perceived that violence occurred because of lack of Consequences (for violent behaviors), lack of Consistency (in security procedures and practices), and lack of Collaboration and support (from management and physicians) while others proposed that the provision of Consequences, Consistency, and Collaboration were required to prevent and/or reduce violent experiences in the ED. The need for Consequences for patients and visitors acting violently toward staff was reiterated by multiple participants including: “Be clear with patient/family/visitors that violence is not O.K., remove (them) from ER with exception of critical illness patient”; “Make it easier for us to ask angry or instigating family and friends to leave”; “Absolute intolerance of verbal and physical violent behavior (is needed)”; and “[Give] Stiffer penalties for offenders, do not let them back in the hospital”. One participant married her plea for consequences with her concerns regarding barriers to reporting violence by saying “We need the patients to be held accountable for their actions. Need management support for our safety – shouldn’t feel afraid to protect yourself, maintain self preservation, in order to protect your employment.” Other participants shared perspectives that helped to frame the need for consequences as a plea for a larger plan to address violent behavior: “Create a ‘care plan’ to refuse care to patients that are frequent offenders, malingers, or to those that threaten staff. Standardize criteria for removal of those threatening staff. Educate staff on policies, [and] encourage staff to take appropriate measures outside of the hospital (i.e. police reports).” The need for Consistency in implementation of policies and consequences was also identified as an essential action to enhance staff safety. Participants’ many pleas for consistent security measures were represented by comments such as “Security is very inconsistent in allowing family back to rooms and limiting the amount of visitors, also inconsistent with checking personal belongings every time someone leaves the ED” and “All visitors need to go thru a metal detector. Visitors need to be restricted to two at a time with no exceptions. All entrances to ED should be locked.” Other issues were raised about consistent implementation of policies “There needs to be a standard policy and procedure on how violent patients [are] to be han-

P. Renker et al. / Staff perspectives of violence in the emergency department

dled. Patient yelling at nurses is OK with some staff and other staff asks patients to leave unless the patient is critical or medical hold. . . there should be a standard way to handle this.” The need for Collaboration was frequently present in participants’ responses. Support from hospital administration was identified as critical for the development, implementation, and enforcement of a zero tolerance policy. The need for and concern related to the lack of a zero tolerance policy was reflected in multiple responses. The perspective that “Management seems to take the side of verbally abusive patients and visitors” and other lack of support from administration was reflected in participants responses to the item “Do you believe your hospital administration and ED management are committed to eliminating workplace violence in the ED?” with “no” responses reported as 68.6% and 60.8%, respectively. Participants also shared that they felt support from hospital administration was essential to “maintain staff levels” and to let them “know my hospital management will back me if a situation turns violent”. In addition, the concerns raised in the conflict theme were also viewed by one participant as a need for physicians and nurses to be “on the same page”. Figure 2 depicts the themes pictorially as a model with contributing factors to the development and maintenance of violence in the ED forming the outer ring. The inner ring encompasses three themes representing cumulative maladaptive reactions to working in an environment characterized by physical violence and verbal abuse: Conflict, Cynicism, and Concern for Customer Service, while the last three themes, Consistency, Consequences, and Collaboration surround the maladaptive reactions. The last themes represent participants’ frustrations with, and desires for, policies and actions to calm the turbulence that violence creates in their professional lives. As such, if positively addressed, the themes of consistency, consequences, and collaboration represent potential protective barriers to mitigate against the development of the maladaptive reactions. Unfortunately, inconsistency and lack of collaboration and consequences can also serve to potentiate the maladaptive reactions of cynicism, conflict, and concern for customer service. In this model, interventions would be focused on developing and strengthening policies and practices that enhance collaboration, consistency, and consequences. 4. Discussion Qualitative research reports differ from quantitative reports in that the meaning and implication of the qual-

15

itative findings are part of the analytic process and reported in the results section while quantitative reports use the discussion section as a framework for discussing the interpretation of the findings. The discussion section in this report, as in other qualitative studies, is used to compare and contrast the study’s findings with those previously published, identify research limitations, and propose areas for future interventions and research. The level of violence reported in the findings from this study can be compared to the results from the two waves of the ENA Longitudinal Surveillance Study on the Prevalence of Violence in Emergency Departments across the United States [10,23]. Two challenges occurred when comparing our sample to the two national ENA samples. The first was that both national studies generalized among the different levels of ED as opposed to reporting findings specific to urban, Level I trauma units such as the one in this study. The second challenge was that the 2011 ENA survey measured the frequency of violence experiences by asking for recall from the last seven days only, while the 2009 survey version (and the one that was used for this study) asked nurses to identify if they had ever experienced any of a list of violent experiences on “each shift worked, weekly, monthly, quarterly, semiannually, less than annually, and never”. Demographically this study’s sample had a lower percentage of individuals from day shifts, had less nursing experience, and was approximately 10 years younger than the national sample. While our study’s findings included similar levels of verbal abuse, our sample’s reports of physical violence were higher than the other two samples. Other commonalities include similar perceptions of the level of safety and adequacy of security, perceptions of support from hospital and ED management, and factors precipitating violence in the ED. Chapman et al. identified ED nurses’ concerns about workplace violence in their qualitative study that related to several of the findings reported for our study including avoiding patients, lack of trust and support, safety concerns, disruption in patient care, and feeling that violence is part of the job [13]. The similarities in the findings between our sample and other studies provide helpful insights. The high level of violence experiences appears to reflect common social influences as opposed to being caused by specific internal and/or external factors. Interventions, therefore, that have been effective in increasing staff safety in other EDs can be considered for use in our ED. Lanza, Zeiss, and Rierdan found that non-physical (verbal abuse) and physical violence are significantly

16

P. Renker et al. / Staff perspectives of violence in the emergency department

modifiable risk factors and address what is in our control including staff training in recognizing escalating anxious or aggressive behavior, policy development and implementation, and environmental changes. It is beyond the scope of any violence prevention program to attempt to change the non-modifiable factors identified as base causes of violence in the ED such as location in an urban area, crime rate, and patient behaviors related to intoxication, drug use, and drug-seeking. The sample in this study reflected their needs for a zerotolerance policy, environmental changes to enhance security screening of visitors and control the traffic flow through the ED, and enhanced security measures, including a need for staff training to recognize and diffuse violent situations. Fig. 2. Model portraying staff perspectives of violence in the emergency department.

5. Strengths and limitations related [30]. Their finding that workers who experienced non-physical violence were more than seven times as likely to experience physical violence as those who had not experienced non-physical violence is highly relevant to our study where every participant identified that they had been verbally abused as it underscores the need to create a workplace environment of staff safety. Complex relationships between customer/consumer/ patient and hospital staff members were highlighted in our findings. While the challenges associated with conflicting staff and end user relationships are not unique to our study or our hospital; they have been intensified with a national emphasis on patient satisfaction (Hospital Consumer Assessment of Healthcare Providers and Systems – HCAHPS) to enhance Value Based Purchasing of services and other compensation strategies for hospital payment [31]. As depicted in our model, interventions that focus on increasing staff safety may remove the angst associated with negative perceptions of “customer service” and return the CARE to patient care that exists independently of the desire to demonstrate high patient satisfaction scores. The purpose of this study was to identify and understand staff concerns so that interventions could be developed to address those concerns and to reduce the level of violence in the ED. In the model representing our research findings for this study (Fig. 2), interventions to increase staff safety and decrease violence would be focused on developing and strengthening policies and practices that enhance collaboration, consistency, and consequences. Interventions developed to minimize violence in the ED must focus on

While the return rate of 37–40% of the survey falls into a generally “acceptable” return rate, it is important to identify that participants returning surveys might have been motivated to do so because of their personal experiences with violence, falsely inflating the prevalence of violence identified above and introducing response bias into the study. It may be relevant to identify that the prevalence of violence in our study is very similar to that identified in two national studies that recruited ED nurses in a comparable cross-sectional approach. However, the same response bias could have been present with the ENA findings even though they were recruited via random sampling techniques. There were insufficient number of ancillary staff to make meaningful comparisons between the prevalence and types of violence experienced as compared to registered nurses. Data from both occupational groups were combined for analysis because of the lack of significant differences in their responses. It is possible that the lack of significant differences was due to a Type II error. The fact that there was only one piece of qualitative data that distinguished any of the comments between nurses and paramedics serves to strengthen the transferability of the qualitative findings. The lack of mutual exclusivity and collective exhaustiveness that occurred with the qualitative data in this study is not the outcome desired by many qualitative researchers [24,26]. For example, several of the codes developed from the data could apply to more than one of the themes. The inability to confine each response to a specific category in this study, however, may reflect the often chaotic events captured when

P. Renker et al. / Staff perspectives of violence in the emergency department

nurses and staff experience violence or abuse in nurse and staff-patient relationships. In addition, the varied nature of the responses, including those not included in the final analysis may reflect on the wide spectrum of participants’ perspectives, experiences, and responses to working in an environment that is characterized by turbulence and uncertainty foretelling that many challenges may occur in attempting to address the needs of all of the ED staff members. The rigor associated with this study is supported by several traditional qualitative approaches. An established data trail encompassing decisions made for thematic analysis and model development was used as well as prolonged engagement of the researcher in terms of receiving feedback about the themes from ED nurses and researchers external to the study’s clinical setting. In addition, positive affirmation from study participants indicating that the findings represented their chief concerns which, qualitatively, is referred to as “grab” and “fit”.

Nurse Manager of the Emergency Department at our hospital at the time this study was conducted. Rick’s perspective of openness and inquiry (related below in his own words) are the reason why our emergency department has the best chance to become a model of safety for staff, patients, and visitors. Rick writes: “Workplace violence” are two words no leader wants to hear, however the dynamics and impact of workplace violence are important for every leader to understand. Violence against healthcare workers is rising and presents challenges for every healthcare leader. Responding to work place violence requires a deeper look into the extent, nature and precipitating factors. The challenge in developing a systematic approach is that the extent of work place violence against nurses in the workplace is not fully understood. It can be a challenge for leaders to find a starting place as we typically have to start after an incident has occurred. There is significant value in surveying staff as a means of gathering information, which will provide direction and a starting place. The survey responses can be difficult for a leader to understand at first, yet it is important to recognize that survey results are information and not a reflection of your leadership.

6. Conclusion The findings from our study provide substantive support for the development of interventions that will hopefully help our staff, patients, and visitors to work in and/or receive care in a safe environment. The model developed to portray the findings from the study will provide a template for the development of policies and actions to increase workplace safety. The findings from this study in relationship to the needs for consistency, collaboration, and consequences represented in the model will serve as the basis for the development of new policies relating to zero-tolerance, reporting violence, security strategies, and administrative response to reports of violence. The findings also help to identify the need to include staff representation with policy development and implementation. After a time period for implementation of the new policies and strategies developed from the study findings, the ED plans to repeat the survey, adding items to reflect staff satisfaction with the interventions. That evaluation will help to test the efficacy of the model developed in this study to address concerns related to workplace safety.

References [1]

[2]

[3]

[4]

Acknowledgments [5]

The authors would like to kindly acknowledge the support of Rick Kelley, BSN, MA, RN, Administrative

17

The Joint Commission [Internet]. [place unknown]: The Joint Commission; 2010. Preventing violence in the health care setting; 2010 Jun 3 [cited 2012 Aug 20]; [3 pages]. Available from: http://www.jointcommission.org/sentinel_event_alert_ issue_45_preventing_violence_in_the_health_care_setting_/. The American College of Emergency Physicians [Internet]. Dallas (TX): The American College of Emergency Physicians; 2011. Protection from physical violence in the emergency department environment; 2011 Jul 12 [cited 2012 Aug 20]; [about 2 screens]. Available from: http://www.acep.org/ Clinical-Practice-Management/Protection-from-PhysicalViolence-in-the-Emergency-Department-Environment/. The Emergency Nurses Association [Internet]. Des Plaines (IL): The Emergency Nurses Association; 2012. ENA Toolkit for Workplace Violence; 2011 [cited 2012 Aug 20]; [about 10 pages]. Available from: http://www.ena.org/IENR/Violence ToolKit/Documents/toolkitpg1.htm. The American Nurses Association [Internet]. Silver Spring (CO): The American Nurses Association; 2012. Preventing Workplace Violence; 2006 Aug 24 [cited 2012 Aug 20]; [2 pages]. Available from: http://nursingworld.org/MainMenu Categories/WorkplaceSafety/bullyingworkplaceviolence/ PreventingWorkplaceViolence.pdf. The United States Department of Labor [Internet]. Washington (DC): The Occupational Safety and Health Administration; 2012 [cited 2012 Aug 20]. Workplace Violence; [1 screen]. Available from: http://www.osha.gov/SLTC/workpla

18

P. Renker et al. / Staff perspectives of violence in the emergency department

ceviolence/. Gates D, Ross CS, McQueen L. Violence against emergency department workers. J Emerg Med. 2006 Oct; 31(3): 331-7. [7] Peek-Asa C, Allaredy V, Valiante D, Blando J, O’Hagan E, Bresnitz E, et al. Workplace violence and prevention in New Jersey hospital emergency departments. The State of New Jersey website [Internet]. 2007 June 19; [cited 2011 Dec 11]; [34 pages]. Available from: http://www.nj.gov/health/surv/docu ments/njhospsec_rpt.pdf. [8] Fernandes CM, Bouthillette F, Raboud JM, Bullock L, Moore CF, Christenson JM, et al. Violence in the emergency department: A survey of healthcare workers. CMAJ. 1999 Nov 16; 161(10): 1245-8. [9] Mayer BW, Smith FB, King CS. Factors associated with victimization of personnel in emergency departments. J Emerg Nurs. 1999 Oct; 25(5): 361-6. [10] Emergency Nurses Association [Internet]. Des Plaines (IL): Emergency Nurses Association Institute for Emergency Nursing Research; c 2012. Emergency Department Violence Surveillance Study; 2011 Nov 2 [cited 2012 Aug 13]; [60 pages]. Available from: http://www.ena.org/IENR/Documen ts/ENAEDVSReportNovember2011.pdf. [11] Kowalenko T, Cunningham R, Sachs C, Gore R, Barata I, Gates D, et al. Workplace violence in emergency medicine: Current knowledge and future directions. J Emerg Med. [Epub 2012 May 24 ; 1-9]. [12] Gates D, Gillespie G, Succop P. Violence against nurses and its impact on stress and productivity. Nurs Econ. 2011 MarApr; 29(2): 59-67. [13] Chapman R, Perry L, Styles I, Combs S. Consequences of workplace violence directed at nurses. Br J Nurs. 2009 Nov 12-25; 18(20): 1256-1361. [14] Kassangra SM, Rao SB, Sullivan AF, Gordon JA, Magid DJ, Kaushal R, et al. A survey of workplace violence across 65 U.S. emergency departments. Acad Emerg Med. 2008 Dec; 15(12): 1268-74. [15] Shields M, Wilkins K. Factors related to on-the-job abuse of nurses by patients. Health Rep. 2009 Jun; 20(2): 1-13. [16] Arnetz J, Arnetz, BB. Violence towards health care staff and possible effects on the quality of care. Soc Sci Med. 2011; 5: 417-27. [17] Farrell GA, Bobrowski C, Bobrowski P. Scoping workplace aggression in nursing: Findings from an Australian study. J Adv Nurs. 2006 Sep; 55(6): 778-87. [18] Roche M, Diers D, Duffield C, Catling-Paull C. Violence toward nurses, the work environment, and patient outcomes. J Nurs Scholarsh. 2010 Mar; 42(1): 13-22.

[19]

[6]

[20]

[21]

[22]

[23]

[24] [25]

[26]

[27]

[28]

[29]

[30]

[31]

U.S. Department of Labor [Internet]. Washington (DC): U.S. Bureau of Labor Statistics; c 2011 [cited 2012 Aug 20, 2012]. Case and demographic characteristics for work-related injuries and illnesses involving days away from work. Available from: http://www.bls.gov/iif/oshcdnew.htm. Child RJ, Mentes JC. Violence against women: the phenomenon of workplace violence against nurses. Issues Ment Health Nurs. 2010 Feb; 31(2): 89-95. Anderson L, Fitzgerald M, Luck L. An integrative literature review of interventions to reduce violence against emergency department nurses. J Clin Nurs. 2010 Sep; 19(17-18): 252030. Kynoch K, Chiung-Jung W, Chang A. Interventions for preventing and managing aggressive patients admitted to an acute hospital setting: A systematic review. Worldviews Evid Nurs. 2011 Jun; 8(2): 76-86. Gacki-Smith J, Juarez A, Boyett L, Homeyer C, Robinson L, MacLean S. Violence against nurses working in U.S. emergency departments. J Nurs Admin. 2009 Jul-Aug; 39(7-8): 340-9. Cresswell J. Research design: Qualitative, quantitative, and mixed method approaches. Thousand Oaks, CA: Sage; 2009. Morse JM, Niehaus L. Combining qualitative and quantitative methods for mixed-method designs. In: Munhall P, editor. Nursing research: a qualitative perspective. 4th ed. Sudbury, MA: Jones and Bartlett Publishers; 2007. pp 541-54. Creswell JW. Qualitative inquiry and research design: Choosing among five traditions. Thousand Oaks, CA: Sage Publications; 1998. Wolcott HF. Transforming qualitative data: Description, analysis, and interpretation. Thousand Oaks, CA: Sage Publications; 1994. Statistical Package for the Social Sciences, inc. Statistical Package for the Social Sciences (SPSS) 17, [software], Chicago, 2008. Wolf ZR. Ethnography, the method. In: Munhall P, editor. Nursing research: A qualitative perspective. 4th ed. Sudbury, MA: Jones and Bartlett Publishers; 2007. pp 293-230. Lanza M, Zeiss R, Rierdan J. Non-physical violence: a risk factor for physical violence in healthcare settings. AAOHN J. 2006 Sep; 54(9): 397-402. Centers for Medicare and Medicaid Services. Washington DC. HCAHPS: Patients’ Perspectives of Care Survey. Downloaded November 8, 2012 from http://www.cms.gov/Medica re/Quality-Initiatives-Patient-Assessment-Instruments/ HospitalQualityInits/HospitalHCAHPS.html.