Emergency department registered nurses overestimate their disaster competency: A cross-sectional study

Background: Major incidents continue to pose a threat to health care systems by overwhelming them with a sudden surge of patients. A major factor impacting a hospital ’ s surge capacity is the skills, abilities, and knowledge of emergency department (ED) registered nurses (RN). The level of disaster nursing competency they possess affects patient safety and outcome. ED RNs ’ ability to accurately assess their competency and knowledge is imperative for mitigating the effect of major incidents. ED RN ’ s perception of overall disaster preparedness has not been thoroughly addressed. The aim of this study was to assess emergency department registered nurses ’ self-perceived disaster preparedness. Method: The study was a cross-sectional study per the STROBE checklist. A self-assessment questionnaire based on the results of a study identifying specific disaster nursing competencies for ED RNs was distributed to all ED RNs at six participating hospitals between January 10th to February 19th of 2019. A five-point Likert-type scale was used to assess competency. Results: ED RNs ’ disaster preparedness according to the Total Disaster Competency mean was low. Furthermore, the results indicate that ED RNs ’ significantly overestimate their disaster nursing competency when compared to the Total Disaster Competency mean. Additionally, this study identified factors such as experience and education were positively associated with disaster preparedness and self-assessment ability. Conclusion: ED RNs ’ overestimate their disaster preparedness. However, ED RNs with experience and education may be better prepared. ED RNs with formal disaster education appeared to have better insight concerning their preparedness. Clinical experience, advanced levels of education, and training were positively associated with preparedness. Overestimating disaster competencies may negatively impact patient outcomes during a major incident.


Background
Major incidents (MI) are omnipresent events that threaten to overwhelm an affected community's health care system by causing a sudden surge of patients in need of health care [1][2][3][4][5].Hospitals play critical roles in providing care during MIs [6,7].Vital strategic measures for mitigating the effects of an MI include increasing surge capacity through the allocation of resources, level of hospital response, and activating alternative, time-effective triage systems [8,9].Hospital's surge capacity is determined by several factors (staff, stuff, structure, and system) [10].In relation to staff, front-line responders, such as emergency department (ED) registered nurses' (RN) skills, knowledge and abilities are essential for providing quality care and minimizing excess morbidity and mortality [11][12][13].A sudden surge of patients to EDs places responsibility on frontline responders often requiring RNs to quickly adapt to meet needs from a wide variety of events as stipulated by the all-hazards approach utilized in disaster response plans [9,14].

Disaster nursing
The International Council of Nurses (ICN) recognizes the importance of nurses' disaster competency in mitigating the effects of MIs, by outlining general disaster core competencies and emphasizing the need for specific disaster core competencies [14,15].Disaster nursing is defined as the ability to apply and adapt the application of competencies to a large variety of events with limited resources to mitigate the effects of an MI [15].RNs constitute the largest group of medical professionals and ED RNs are among the first to receive, assess, and treat victims from a major incident.Thus, underscoring ED RNs' role in patient safety [4,16].It is generally recognized that ED RNs' disaster competencies, during and after a major incident are crucial [15,17].ED RNs' ability to accurately assess their competency, knowledge gaps and needs is imperative for patient outcomes [18].
National doctrines and health organizations stipulate the need to both develop disaster medicine competencies for a broad range of possible incidents and evaluating preparedness [19,20].Despite its importance, little attention has been given to evaluating ED RNs' disaster nursing competencies.Evidence supports that self-assessments, although subjective, are valid measurements of competency and are often used to evaluate RN clinical competency [18,21].Previous studies employing self-assessment instruments to evaluate RNs' disaster preparedness have reported moderate to low levels of disaster preparedness [22][23][24][25][26][27][28][29].However, ED RNs' perception of overall disaster preparedness has not been addressed.There are to our knowledge, no prior studies assessing ED RNs' perceptions of their disaster competencies in relation to disaster preparedness and as measured by an instrument constructed specifically for ED RN's.
The aim of the current study was to assess emergency department registered nurses' self-perceived disaster preparedness.

Study design
The study used a validated questionnaire to conduct a cross-sectional study (in accordance with the STROBE checklist) using descriptive and inferential statistics.

Participants and setting
All seven major EDs in the region of Stockholm, Sweden were invited to participate.Six accepted.The study period was January 10th to February 19th of 2019.Inclusion criteria were all ED RNs employed at the respective Eds.Nurses employed by independent staffing agencies were excluded due to them not having an email address connected to the hospital.
It is noteworthy to acknowledge that in the study setting the term "major incident" is used in regards to events and response while "disaster" is used when referring the area of research and an events affect on a society.

Ethical consideration
Ethics approval was obtained by the regional ethical review authority (diary number 2018/2142-31).Each participant was provided a description of the study and informed that participation was voluntary, and anonymity assured.In addition, participants were informed that they could withdraw from the study at any time.

The questionnaire
The basis of the instrument in this study is a previously validated instrument used to assess RN disaster competencies, the Emergency preparedness instrument questionnaire (EPIQ) [25].The EPIQ is a highly reliable (α values ranging from 0.92 to 0.98) and valid instrument [24,28,29].Prior to conducting the current study, essential ED RN disaster competencies were identified using a modified Delphi technique based on the EPIQ instrument [30].The instrument was modified initially modified through expert consensus [30].The number of items was increased through item refinement, ie expanding items that convey more than one idea, or "double-barreled" items [31].In addition, terminology specific to the study setting as well as items concerning legislation specific to the setting were included [30].In addition, the scale for self-assessment was refined.To minimize lexical ambiguity and increase participants mutual interpretation of the scale used, each step in the five-point Likert-type scale used in this study was clearly defined with each step corresponding to Benner's stages of clinical competence (1 = Novice, 2 = Advanced beginner, 3 = Competent, 4 = Proficient, 5 = Expert [32]).The scale and corresponding values were clearly explained and defined in the questionnaire.
A pilot study was conducted with RNs enrolled in emergency nursing and ambulance masters' programs at Sophiahemmet University to assess comprehension of the items, understanding of the Likert-type scale used as well as to assess the amount of time required to complete the questionnaire.A total of 15 questionnaires were completed.The questionnaire took between 7 and 20 min to complete with participants suggesting an adjustment of some of the wording of the items to aid in clarity.The authors analyzed feedback and edited items to increase understanding.
The Delphi study and pilot study resulted in the instrument used in this this study, which consisted of nine items concerning general background, and 60 self-assessment items divided into 12 dimensions.These refinements of the instrument may have increased its reliability.Internal reliability for this questionnaire as expressed by Cronbach's alpha was α = 0.989 (see Supplementary file Table 1).

Reduction of items and dimensions (Exploratory factor analysis)
An exploratory factor analysis (EFA) with a threshold of 0.5 was conducted upon data collection resulting in a reduction of items (n = 46) and dimensions (n = 3).After analysis of the items in each dimension, new names for the dimensions were decided on by the authors.The instrument used for analysis consisted of 28 items relating to "Staff, Stuff, Structure, System", 10 items relating to Chemical, Biological, Radiological, and Nuclear (CBRN) 8 items relating to "Specific patient groups" and one question asking participants to rate their overall disaster preparedness (Supplementary file 1 Table 1).

Data collection
An email with information detailing the study and a link to the questionnaire was sent to hospital managers for distribution.Hospital managers then mailed the information and link to ED RNs.Data collection was completed using a closed link to Google Forms® which was emailed to the participants.The questionnaire was closed two weeks after a third reminder.

Data analysis
Data from the online questionnaire was first transferred to Micro-soft® Excel® for Office 365, coded and then imported to IMB® SPSS® version 26.0 for analysis and JASP version 0.9.2 (JASP Team 2018).To explore possible underlying relationships of items, or variances between items measuring competency, an exploratory factor analysis was conducted [33].Internal reliability of the instrument was assessed using Cronbach's α.Missing values.Non-missing values were used for analysis.In addition to descriptive statistics, the Mann-Whitney u-Test, Kruskal-Wallis, were used to assess differences between groups and means.Correlation analysis was conducted using Spearman's tau-b correlation.
The response alternative "uncertain" was treated as user missing data resulting in dichotomization of prior MI response experience, disaster medicine education at the bachelor's level, and education level (bachelor and advanced).This also reduced the number of categories for "frequency of training" from five to four.

Description of the participants
Seven hospitals were invited to participate: one declined.The study population according to ED management was comprised of a total of 372 RNs employed at the six participating EDs.A total of 140 nurses completed the questionnaire resulting in a response rate of 38% (n = 140/372).The majority of the nurses (70.4%) were between the ages of 26-44.65% percent of nurses had more than five years of nursing experience.40% percent of the participants reported having prior major incident experience.30% of the RNs had advanced degrees within a variety of specialties.In addition, 54.6% of RNs had disaster medicine as a part of their bachelor's curriculum (Table 1).

Nurses' disaster competency
A composite score of the three dimensions was calculated described as Total Disaster Competency.The mean for Total Disaster Competency was 2.34.Means for the three subdimensions were 2.89 (Staff, Stuff, Structure, System), 2.00 (CBRN), and 2.17 (Specific patient groups) (Table 4).The final item assessing nurses' perception of their overall disaster preparedness was statistically significantly higher (M = 2.74) than the Total Disaster Competency (M = 2.34, ρ = 0.000) (Table 2).
There were significant differences in means for all dimensions based on several factors as exemplified by the means for Total Disaster Competency; Means increased based on the level of education (bachelor's degree M = 2.03 advanced degree M = 3.07 ρ = 0.000), clinical experience, (1-3 years M = 1.67, over 10 years M = 2.99 ρ = 0.000), being an instructor (M = 2.98 vs not being an instructor 2.13 ρ = 0.000) having prior MI experience (M = 2.68 vs M = 2.08), and having formal disaster medicine education vs not having disaster medicine education (M = 2.10 vs. 2.61 ρ = 0.000).There were no significant differences based on gender.(Table 3).
These same factors were significantly correlated with RNs' perception of their overall disaster preparedness.The mean score of their perceived preparedness tended to increase with professional clinical experience (r = 0.623), level of education (r = 0.470), and prior MI experience (r = 0.373), with the exception being for formal disaster medicine education, defined as a structured course as part of RNs' bachelor degree, [34] (Table 1) which decreased (r = − 0.293).RNs without formal disaster medicine education assessed their preparedness significantly higher than RNs with formal disaster medicine education (M = 3.08 vs 2.41 ρ = 0.004).

Discussion
The main result of the current study was that all RNs overestimated their overall preparedness for working during a major incident.RN's perception of their preparedness was "less than competent" according to Benner's stages of clinical competence.However, this was significantly higher than the Total Disaster Competency score which was closer to "advanced beginner" than "competent".

Self-assessment of competency for rare events may impact evaluation
Means for the three subdimensions indicate that disaster preparedness was "less the competent".A possible explanation for the low means may be that ED RNs rarely are exposed to certain situations or medical conditions.For example, many of the components that make up the subdimension with the highest mean (2.89) "Staff, Stuff, Structure, System" may more closely mirror many of the competencies ED RNs incorporate during normal circumstances.In contrast, competencies related to infrequent events such as chemical spills, pandemics, biological, and radiological events were lower (2.00) "CBRN" than the other two subdimensions.
Similarly, many ED RNs in the current study have limited contact with pediatric patients.This may explain the lower mean (2.17) for the subdimension "specific patient groups".While RNs in this study may be aware that they lack pediatric competency, this may also indicate both a discrepancy between experience and knowledge.For example, under  normal circumstances, pediatric patients are referred to specific hospitals for care, which minimizes nurses' contact with these age groups.However, during an MI, regular routines may be circumvented.Previous studies have indicated that hospitals closest to the incident site may receive a majority of patients from the incident.ED RNs may receive patients of various ages and medical conditions within minutes of an incident and often before hospitals have been alerted [3][4][5].This may further highlight a gap between knowledge and experience.
According to the results in the current study, ED RNs may erroneously assume that normal standards and regulations apply under extraordinary circumstances, possibly negatively affecting patient outcomes.
Further affecting self-assessment in this study, is that RNs were asked to assess their competency for MIs, events that the majority had little or no experience with.This may lead to RNs neglecting to account for factors such as stress when assessing abilities [35] and increase the likelihood of overestimating abilities.In addition overestimating abilities awareness may impede an active pursuit to fill knowledge gaps, improve or maintain necessary skills, which may negatively impact patient outcome during an MI [36][37][38].

Factors associated with ED RN preparedness
While competency was low, in line with previous studies, several factors may improve preparedness [26,27].Clinical experience and prior MI experience were also strongly correlated to disaster  preparedness, while level of education, and being a trainer were moderately correlated.Both clinical experience and MI experience are factors that are difficult for disaster preparedness coordinators to affect.
To respond to this challenge, there is a need for quality educational methods.Interventions aimed at increasing formal competency through education and training may have a significant impact on nurses' disaster preparedness as evidenced by these results.While having an advanced degree was positively correlated to higher levels of competency, having formal disaster medicine education was negatively associated with competency.This apparent paradox may be explained by the Dunning-Kruger affect in which participants may overestimate their abilities due to a lack of adequate awareness [39].Yet another explanation may be that those with higher levels of expertise more accurately assess their abilities [18].Their perception more closely mirrored reality, reflecting previous studies correlating accurate selfassessments with higher levels of expertise [18].
Improved ED RN disaster preparedness may be achieved through frequent exercises and interactive training where skills and knowledge are evaluated.

Limitations
The validity of self-assessments is an oft-discussed topic with a particular focus on the correlation between self-assessments and actual ability [35].There is however evidence suggesting that self-assessments may be valid measurements of ability, correlating accuracy of selfassessments with the participants' level of expertise [18,21].
Another factor affecting the validity of self-assessments is the risk of participants' subjective interpretation of questions or terms used to measure competency [40].Many studies apply Likert-type scales to assess self-evaluation of preparedness using measurements such as "how familiar" or "to what degree" [22,24].Lexical ambiguity of key terms or lack of participants' mutual interpretation of the scale used may inhibit accurate measurements and negatively affect the reliability and validity of the results.However, the current study attempted to reduce lexical ambiguity by using a scale of measurement that was clearly defined in the questionnaire.By clearly correlating the five-point Likert-type with the five stages of competence corresponding with a five-point Likert scale [32] lexical ambiguity may have been minimized and RNs' mutual interpretation of the scale used increased, thereby increasing the reliability and validity.
Cronbach's alpha was 0.989 and while relatively high, was similar to previous studies [24][25][26]29].Alpha values > 0.90 may may also indicate redundancy of items [31,41].However, refinement of the instrument including item refinement may explain the high alpha value and indicate that this instrument is a psychometrically sound questionnaire for evaluating RNs' self-assessed disaster preparedness.
Due to the response rate, the generalizability of the results should be done with caution.However, the response rate is similar to prior studies [26][27][28] and may be applied to similar settings and contexts.

Conclusions
The current study indicates that ED RNs rate their disaster competencies slightly lower than "competent" and that this self-assessment may be an overestimation of actual preparedness indicating that preparedness may be lower than the results show.However, RNs with formal disaster medicine education may have a more realistic view of their competencies and preparedness.While experience, training, and education correlated with disaster preparedness, the results indicate that ED RNs with advanced degrees may be better prepared for the challenges MIs present.

Relevance to clinical practice
The clinical relevance of this study emphasizes that patient safety and outcome during an MI could be negatively impacted due to ED RNs overestimating their disaster competencies in addition to having low levels of preparedness.Furthermore, ED RNs may lack insight concerning their disaster competencies, which may impede an active pursuit of knowledge and skills needed to improve competencies.These results could aid nursing leaders and disaster preparedness coordinators in disaster training and evaluation of ED RNs' disaster preparedness.

Table 1
Description of the participants n = 140.
*Re-coded as user missing data for analysis ** Formal education course.Systematic, structured, and institutionalized course at the university level.

Table 2
Means per dimension.

Table 3
Means and standard deviations for dimensions and factors.

Table 4
Correlation of demographic factors and disaster competencies.
J.P.Murphy et al.