key: cord-1046235-qxqlhxpg
authors: Lopez, Jamie; Bindler, Ross J.; Lee, Jillian
title: Cross-sectional Analysis of Burnout, Secondary Traumatic Stress, and Compassion Satisfaction among Emergency Department Nurses in Southern California Working through the COVID-19 Pandemic.
date: 2022-03-28
journal: J Emerg Nurs
DOI: 10.1016/j.jen.2022.03.008
sha: 78e6c45b4165c05fd7bbeae5cb0d0a9727ea979f
doc_id: 1046235
cord_uid: qxqlhxpg

OBJECTIVE: The purpose of this study was to assess burnout, secondary traumatic stress, and compassion satisfaction scores in emergency nurses after working through the coronavirus disease 2019 (COVID-19) pandemic using the Professional Quality of Life Scale version 5 (ProQOL) and compare those scores to similar studies conducted pre-pandemic. METHODS: A cross-sectional analysis of a descriptive survey including the ProQOL questionnaire was sent to nurses working in the emergency department prior to 2021 from urban, adult, and pediatric receiving hospitals in Southern California. Results were analyzed to provide insight into the effect of the COVID-19 pandemic on the levels of burnout, secondary traumatic stress and compassion satisfaction compared to pre-pandemic studies found in the literature using the same ProQOL instrument. RESULTS: Mean sub-category scores were in the moderate range for burnout (25.6), secondary traumatic stress (24.5) and compassion satisfaction (38.7). Burnout scores for mid-shift nurses were found to be significantly higher than day-shift nurses (mean difference=5; p=0.024) as were secondary traumatic stress scores (mean difference=4.6; p=0.007). Also, compassion satisfaction sub-category scores in nurses with one child were significantly higher than those with two (mean difference=6.7; p=0.025). DISCUSSION: The unnormalized mean findings were similar to pre-pandemic studies conducted using the same ProQOL instrument indicating nurses are at risk for compassion fatigue. Additionally, the scores from mid-shift nurses reflect increased burnout and secondary traumatic stress while nurses with two children had lower compassion satisfaction. This implies the need for leadership to proactively seek interventions to support nurses on each shift.

Contribution to Emergency Nursing Practice 55 56

Emergency nurses are at risk for burnout and compassion fatigue due to the nature of the 57 environment, variety of patient presentations, and high acuities 24-hours a day. The COVID-19 58 pandemic created uncertainty, changes in practice, and a lack of resources for healthcare 59

providers. 60 61

The main findings of this paper are current unnormalized burnout and secondary traumatic stress 62 scores for emergency department nurses are moderately high while compassion satisfaction is 63

moderately low yet not significantly different from pre-pandemic levels reported in similar 64 studies. Nurses who work the mid-shift scored significantly higher for burnout and secondary 65 traumatic stress than those working the day shift. While those with two children scored 66 significantly lower for compassion satisfaction than those with one child. 67 68

Emergency department nurses need continued support from those in leadership positions to 69 decrease burnout and secondary traumatic stress while continually supporting compassion 70

satisfaction. Department heads must support efforts to decrease negative components of nurses' 71

professional quality of life with specific interventions. The findings suggest the pandemic may 72

have mimicked the status quo. Professional quality of life, as represented by the constructs examined using the 80

Professional Quality of Life Scale version 5 (ProQOL), is made up of both positive and negative 81

emotions that an individual feels while engaged in their position. 1 Emergency nurses experience 82 unique daily stressors, including a multitude of patient presentations, traumatic events, chaotic 83 environments, and a wide range of emotional encounters. 2 Nursing is emotionally, spiritually, 84

and physically demanding; these demands and stressful encounters can lead to compassion 85 fatigue, burnout, secondary traumatic stress, vicarious trauma, and a decrease in compassion 86

satisfaction. 2-6 The ProQOL is a validated tool that measures two negative (burnout and 87 secondary traumatic stress) and one positive (compassion satisfaction) affects, or 88

feelings/emotions, that are part of a worker's overall professional quality of life. 7 89

The ProQOL defines burnout as one of the multiple elements in compassion fatigue. 7 90

Compassion fatigue and burnout have similar symptoms; however, they differ in cause.

Compassion fatigue occurs due to the element of caring for people in a helping profession, while 92 burn out is caused by workplace environments and is associated with the organization rather than 93 the patients. 11 If caregiver's stressors are caused by work conditions, scheduling conflicts, 94

poor work environment, and leadership, this would classify as burnout. 9, 11 Compassion 95 fatigue occurs subtly and is progressive in nature. The signs may go unnoticed until the caregiver 96

suffers from a sudden onset of physical, emotional, and spiritual exhaustion. 8, 9 The 97

consequences of compassion fatigue is the nurse's ability to show and feel compassion is 98 decreased or absent. Nurses who are experiencing compassion fatigue may negatively impact 99 patient outcomes and patient satisfaction due to their emotional distress, inability to care, 100

and low levels of efficiency. Compassion fatigue can be devastating to the nurse both personally 101 and professionally, and early recognition and prevention is important. 10, 11 102

Secondary traumatic stress is a consequence of caring for people who have experienced 103 trauma even though the nurse has not experienced the trauma personally. 10, 11 Secondary 104 traumatic stress describes various symptoms similar to those of post-traumatic stress disorder 105

(PTSD) acquired through secondary exposure to trauma rather than a direct event. 12 Nurses with 106 secondary traumatic stress experience similar symptoms of post-traumatic stress disorder and 107

include behavioral components such as, irritability, anger, lack of sleep, and 108 substance abuse problems. 11 Although secondary traumatic stress and vicarious trauma are often 109 used interchangeably, they were independently developed by different research groups. While 110 secondary traumatic stress is defined by symptoms, vicarious trauma tends to focus on the 111 potentially harmful changes that can occur in professionals dealing with graphic and/or traumatic 112 material through clients and/or patients. 13 

Finally, compassion satisfaction is what nurses' look to achieve by working with 114

people. Many nurses enter the field to help people and improve patient's quality of life. 115

Compassion satisfaction is the knowledge that the caregiver's work is making a positive impact 116 on the patient's experience and meaningful connections are made. 14 117

Immediately after the World Health Organization declared the coronavirus disease 2019 118 (COVID-19) a global pandemic on March 11 th , 2020, followed shortly after by the United States  119 issuing a national emergency proclamation on March 13 th , 2020, the professional demands of 120 emergency nurses were impacted. 15-17 COVID-19, caused by the severe acute respiratory 121 J o u r n a l P r e -p r o o f syndrome coronavirus 2 (SARS-CoV-2), has caused additional stress, uncertainty, and fear in the 122 health professional community particularly including emergency nursing staff. 18-20 When nurses 123 are exposed to unpredictable changes and stressful situations on a regular basis, they can suffer 124 from anxiety and exhaustion, which can lead to decreases in job satisfaction and the quality of 125 care they are able to provide. 21 126

The concepts of compassion fatigue, burnout, secondary traumatic stress, vicarious 127 trauma, and compassion satisfaction have been studied extensively in an array of settings 128

including nursing departments. The overall goal of this project was to record and assess levels of 129 burn out, secondary traumatic stress, and compassion satisfaction using the ProQOL V 130 questionnaire in emergency nurses employed in Southern California during the COVID-19 131

pandemic. The specific aims of this study were to: 132 1) Describe the sample population of emergency nurses 133

2) Summarize the ProQOL V scores by sub-categories (burn out, secondary traumatic 134 stress, and compassion satisfaction) 135

3) Examine correlational relationships between the sub-category scores 136 4) Evaluate ProQOL V sub-category scores by demographic, socio-, work-related 137 characteristic compassion satisfaction 138 139

Methods: 140 141

The study was designed as a descriptive observational study for emergency department 142 nurses at three Southern California acute care hospitals located in Torrance, Laguna Beach, and 143

Mission Viejo where the authors of this article are employed. The Registered Nurses at the 144 selected locations who met inclusion criteria were invited to participate. The inclusion criteria 145

were any registered nurse who was actively working on the unit and hired prior to January 2021.

Qualified individuals were sent an anonymous survey to be completed via the organization's 147

Research Electronic Data Capture (REDCap) instance. 22, 23 Prior to initiation of procedures, the 148 study was reviewed by the organization's regional Institutional Review Board (IRB) as well as 149 the research department and local administrative staff. Due to the anonymous responses and lack 150 of intervention as well as follow-up, the IRB deemed the project exempt as it did not qualify as 151 human subjects research, however all system policies and procedures as well as good research 152 practices were followed. Due to the cross-sectional, hypothesis generation nature of this project, 153 no power or sample size calculation was completed.

Survey Tools 156

The data collection survey included multiple demographic, socio-, and work-related items: 157

i. Years of experience as a nurse, 158

ii. Full-time employment status, 159

iii in health care workers. 7 Although areas of concern with the scale's validity and reliability in 170

nurses have been recently documented, there is a documented lack of more adequate tools. 24, 25 171

The scale is made up of 30 items which are "I"-statements regarding professional work that the 172 user responses to via a 5-point Likert scale ranging from 1 = never to 5 = very often. 7 Scoring is 173 completed by summing the values for all statements in each of the three subscales: compassion 174 satisfaction (10-items), burn out (10-items), and secondary traumatic stress (10-items). 7 It is 175 important to note that when scoring the proQOL V, multiple burn out items are 176 scored inversely. 7 Raw, or unnormalized scores, for each sub-category can be evaluated or 177

transformed to a t-score where the sample mean is converted to 50 with the standard deviation 178 set at 10. 7 179 180

Procedure 181

The survey was distributed via email, which included an attached flyer with a scannable 182 QR code; the recruitment flyer was also posted in participating locations' breakrooms. In the 183 distribution phase, the nurses were informed as to why this survey was being conducted. The 184

aims of the study were explicitly described to the nurses as an assessment of the baseline levels 185

of burn out, compassion fatigue, compassion satisfaction, and secondary traumatic stress in order 186

to not only compare the scores to pre-pandemic levels in similar studies, but also to have 187 knowledge for future interventions designed to help the staff and prevent burn out and 188 compassion fatigue. Since the ProQOL V is a psychological test, it was important to inform the 189 nurses that they were not being evaluated for any bad behavior or feelings and the responses 190 would be anonymous. For description of the study's participant population all responding nurses were pooled 199 and demographic, socio-, and work-related items for the entire study population were described 200 via frequency reporting. For analysis of each of the three ProQOL V sub-category (burn out, 201 secondary traumatic stress, and compassion satisfaction) mean scores and standard deviations 202

were calculated via results from all responding nurses; the mean scores were evaluated directly 203 by the guidelines in the ProQOL manual. All statistical analysis conducted had an alpha set 204 below, 0.05 as the cut for significance. Bivariate correlations between all of the sub-category 205 scores were conducted and significant interactions reported using Pearson's Correlation 206

Coefficient (PCC). Using demographic, socio-, and work-related characteristics as grouping 207

factors, mean ProQOL V sub-category scores were compared using a one-way analysis of 208 variance (ANOVA) to assess differences between all groups. Due to the unbalanced groups and 209 exploratory nature of this post-hoc analysis, any result with an alpha of less than 0.1 was further 210

examined via a Tukey test to examine potential between group differences in mean sub-category 211

scores.

Setting and Context 214 The survey was sent out to three hospitals with the concentration of responses coming 230

from Mission Viejo and Torrance. Responders tended to be older in age (82% were 30 years or 231 above) and experience (82% had at least three years of experience) as well as full-time workers 232

(88%). The reported shifts being worked was split with day shifters being the most common 233 responders (46%) followed by night shift (34%) and mid shift (20%). The majority of the study 234 population were married (60%) and had at least one child (60%). Full demographic 235

characteristics of the responders are further detailed in Table 1 . 236

The mean raw (unnormalized) ProQOL V scores for all three sub-categories were found 237

to be in the "moderate" range: burn out=25.6 (sd=5.6); secondary traumatic stress=24.5 (5.4); 238 compassion satisfaction=38.7 (sd=5.4). Prior to additional analyses, normality of all three sub-239 category results were evaluated and found to be acceptable (Table 2) . Correlations between the 240 three ProQOL V sub-categories were assessed, and responders' burn out scores were found to be 241 associated with both the secondary traumatic stress and compassion satisfaction scores 242

(Supplemental Table 1 ). Burn out and secondary traumatic stress had a significant 243 positive correlation indicating that as the score for one increased, the score for the other also 244 increased (PCC=0.54; p<0.001). While burn out and compassion satisfaction scores were 245

significantly inversely correlated with one-another revealing as one score increased the other 246 score tended to decrease (PCC=-0.64; p<0.001). There was no significant relationship between 247 secondary traumatic stress and compassion satisfaction found. 248

After grouping responders by the demographic characteristics described in California, along with short staffing, changes in practice, critically ill patients and overcrowded 290 departments. Figure 1 and Supplemental Table 2 illustrate the current study's results along with 291 those that have been previously published with a unnormalized measure of central tendency 292

(point estimate) and variability (standard deviation). 293

Although a full systematic review or meta-analysis was not conducted, a targeted 294 literature review focusing on examining the potential impact of the COVID-19 pandemic on 295 results was conducted using a Preferred Reporting Items for Systematic Reviews and Meta-296

Analyses (PRISMA)-like process. 28 . After completion of a literature review, three manuscripts 297 met all requirements and were included: 298

1. "The effectiveness of an educational program on preventing and treating 299 compassion fatigue in emergency nurses", 29 300

2. "Factors that influence the development of compassion fatigue, burnout, and 301 compassion satisfaction in emergency department nurses", 30 and 302

3. "Differences in Compassion Satisfaction, Compassion Fatigue, and Work 303

Environment Factors by Hospital Registered Nurse Type". 31 

Some areas of consideration for this lack of major change from pre-pandemic times may 306

be that emergency nurses have always adapted to rapid changes in patient acuity and high levels 307 J o u r n a l P r e -p r o o f of stress, thus are more resilient to the environment created by the pandemic. Another possibility 308

is the increased support from the leadership, the organization, and the public during the 309 pandemic. Nurses were recognized as heroes during the early pandemic which may have created 310 a sense of support that prevented severe rises in burn out and secondary traumatic stress scores. 311

Also, the study took place during a period when COVID-19 cases were relatively low (after 312 public distribution of COVID-19 vaccines began, and before the major wave of cases caused by 313 the delta-variant began). As by July 27 th , 2021, the seven-day case average was over 60,000 314 cases per day compared to the 20,338 case per day average seen during the study period. 32 315

While the mean scores were not different from earlier studies, certain demographic 316 characteristics appear to have an impact on sub-category scores. When grouped by the shift 317

worked, dayshift, mid-shift, and night shift, the mid-shift nurses have significantly higher levels 318

of burn out and secondary traumatic stress. This finding provides insight for the need to support 319 the mid-shift nurses and further investigate what contributes to this result. Mid-shift nurses start 320 the shift during busier times and end the shift during busier times. They often miss in-321

services scheduled for the morning and night shifts. Their break times do not fall at usual break 322

times which may prevent participation in the hospital's scheduled nursing events. Working the 323 mid-shift may lead to missing department gatherings, supportive interventions, and pre-shift 324

informational meetings and huddles. Plus, when the mid-shift nurses arrive, they are often sent 325 straight out to the unit to take over an assignment, give lunch breaks, or relieve an overwhelmed 326

co-worker. 327

The number of children appears to have an effect on the compassion satisfaction sub-328 category score. Nurses with one child had significantly higher compassion satisfaction scores 329 than those with two. The average compassion satisfaction score for those with one child nearly 330 improved from the "moderate" range into the "high" range. Although emergency nursing leaders 331

do not have a direct impact on the number of children an individual has, additional support, such 332

as flexible scheduling and non-punitive available time off to care for family can be provided to 333 lessen the impact of child-care stressors on their professional quality of life. 334

The implications of the results in this current study compared with previous studies 335

indicate that nurses are constantly at risk for the development of compassion fatigue and early 336 recognition and continuous interventions are needed. When nurses experience burnout and 337 secondary traumatic stress combined with low levels of compassion satisfaction they will slowly 338 and quietly develop compassion fatigue. Once the nurse is experiencing compassion fatigue a 339 sudden onset of physical and emotion exhaustion may occur and the ability to recover becomes 340 difficult. 11 14 Patient outcomes suffer, turnover rates increase, unit moral decreases, and the 341 nurse's mental health declines. The consequences are devastating to the nurse, the patients, and 342 the organization. 14 343 344 345

Limitations 346 347

The study had several limitations. First, the survey was sent out in May 2021, which was 348 at a time where the COVID-19 surge in January had decreased and work life had returned to a 349 more normal state. Second, the study's sample (N=50) was limited to a single hospital system in 350

Southern California. Also chosen not to participate. Finally, the demographics did not include sex or education level. 356

The ProQOL V manual explains that sex is not a significant factor in different scores; however, 357 looking at the results obtained, these demographics may provide additional insight. 7 For 358 example, were the nurses with two children who scored low on compassion satisfaction male or 359 female? Additionally, do the nurses working mid shift voluntarily work those hours or are they 360

waiting for a dayshift spot to open. Nurses with children may work mid-shift for childcare 361 reasons yet the hours make sleeping and self-care difficult. These factors may contribute to high 362 burn out and low compassion satisfaction and not be due to the organization. Further investigation 363

into variations of the ProQOL V score based on education level (ADN, BSN, MSN) , certification 364 such as Certified Emergency Nurses (CEN) and any education on self-care and/or emotional 365

intelligence could be explored on their potential effect on the nurses' scores.

As with all observational studies the risk of bias is present. The small sample size and 367

limited locations of participants as well as the fact that generally motivated individuals complete 368

optional studies all play into the risk of selection bias. 33 Due to the potential risk, generalization 369

of the results of this study applying to all emergency nurses throughout the US, or beyond, 370

without the application of further analysis would not be advisable.

Implications for Emergency Clinical Care 373 374

An important point in this study is that while the ProQOL V scores of the emergency 375 nurses are not significantly higher in burn out and secondary traumatic stress or significantly 376 lower in compassion satisfaction from the pre-pandemic scores found in similar studies, they are 377 still in the moderate range as the point-estimate for compassion satisfaction being lower than 378

previous findings and burn out and secondary traumatic stress point-estimates being higher.

Continuous work to improve these scores and the emergency nurses' quality of life is 380 needed. Those in leadership positions need to be aware of the increased burn out and secondary 381 traumatic stress levels in the mid-shift nurses and should include the mid-shift nurses in all 382 supportive interventions. 383

Interventions to support higher compassion satisfaction levels and decrease burn out and 384 secondary traumatic stress levels may cultivate a better work environment for our nursing 385 professionals, increase patient satisfaction, and decrease turnover. Interventions which focus on 386 nurse recognition have shown to be effective in reducing burn out and turnover. 34 Cost effective 387

interventions can be implemented as part of a unit-based improvement project. Items such as a 388

gratitude board in the breakroom where staff can write a thank you message openly to each 389 other, anonymous suggestion boxes which are addressed in unit meetings, leadership rounding 390

where leaders show support and interest in staff, and a time and place for staff to meet, debrief, 391

and provide feedback are shown to be effective. 34 392

Compassion fatigue prevention in nurses can be targeted by interventions based on self-393 care and self-awareness. Activities such as meditation, mindfulness exercises, journaling, 394 adequate sleep, healthy diet, and exercise are self-care activities that support well-being. Some 395 hospitals have a dedicated self-renewal room where nurses can relax in times of stress. Other 396 hospitals have used mobile carts with refreshments, journals, aroma therapy, stress reduction 397 balls, and portable chair massagers. 35 Leadership can support nurses by implementing proactive 398 programs targeted at improving nurses' well-being. 14 Leaders can also provide education on the 399 J o u r n a l P r e -p r o o f concepts of burn out and compassion fatigue and the importance of self-awareness and early 400 recognition in order to prevent compassion fatigue. 14 Nurses may not be aware of what 401 compassion fatigue is and how it starts. Simple education on the concept to promote awareness is 402

an important for prevention. The efforts will benefit the nurse, the patient, and the organization.

Conclusion

The results of this study show that emergency nurses had moderate levels of burn out and 407 secondary traumatic stress with moderate levels of compassion satisfaction indicated probable 408 compassion fatigue while working through the COVID-19 pandemic. The present study's results 409

were not significantly different from the pre-pandemic levels reported in other nursing 410 studies. Emergency nurses are accustomed to unexpected events on a daily basis. They 411 work in chaotic environments and adapt to changes throughout the shift based on incoming 412

patients. This research did not receive any specific grant from funding agencies in the public, commercial, 443

or not-for-profit sectors.

Acknowledgments 446

The authors would like to thank the nursing professionals who participated in the survey and Dr. 447

Trisha Saul for guidance and support throughout the process. Standard deviation (SD) 482 

Vicarious traumatization: an empirical study of the effects of 576 trauma work on trauma therapist secondary traumatic stress. Professional Psychology: Research 577 and Practice

583 15. World Health Organization. WHO Director-General's Opening Remarks at the Media 584 Briefing on COVID-19. World Health Organization Director-General

WHO Declares COVID-19 a Pandemic

COVID-19) Pandemic. The White House

Implications for COVID-19: A Systematic Review of Nurses' Experiences of Working in Acute 599

Care Hospital Settings during a Respiratory Pandemic

COVID-19: Impact on Nurses and Nursing

Impact of COVID-19 Outbreak on Nurses' Mental 606 Health: A Prospective Cohort Study

Compassion Fatigue among healthcare, emergency, and community 610 service workers: A systematic review

Research electronic data 614 capture (REDCap) -a metadata-driven methodology and workflow process for providing 615 translational research information compassion satisfaction support

The REDCap consortium: building an international community 620 of software platform partners

The ProQOL-21: a revised 623 version of the Professional Quality of Life (ProQOL) scale based on Rasch analysis

Psychometric Evaluation of the ProQOL Version 5 for Assessing Compassion 627

Burnout and Secondary Traumatic Stress in Nurses

Centers for Disease Control and Prevention. Trends in Number of COVID-19 Cases and 634 Deaths in the US Reported to CDC, by State/Territory

The Effectiveness of an Education program on 641 Preventing and treating Compassion Fatigue in Emergency Nurses

Factors that Influence the Development of 645

Compassion Fatigue, Burnout, and Compassion Satisfaction in Emergency Department Nurses

Satisfaction, Compassion Fatigue, and Work Environment Factors by Hospital Registered Nurse 650 Type

Articles-Previous-Topicompassion 653 satisfaction/Differences-in-Compassion-Satisfaction-Compassion-Fatigue-and-Work-654 Environment-Factors-by-Hospital.html 655 656 32. Centers for Disease Control and Prevention. Delta variant: what we know about the science

Bias in observational studies. Statistics compassion satisfaction and Research 661 Design

The Implementation of a cultural change toolkit to 664 reduce nursing burnout and mitigate nurse turnover in the emergency department

Code Compassion: A caring fatigue reduction intervention

Cross-sectional Analysis of Burnout, Secondary Traumatic Stress, and Compassion Satisfaction 714 among Emergency Department Nurses in Southern California Working through the COVID-19

Search methods for identifying previously published raw, unnormalized, ProQOL V sub-723 category scores (and standard deviations) for United States-based nurses

Scholar using key search terms in various combinations: 727 a. Key search terms: 728 i

• Articles were restricted to peer-reviewed journals

• Only sub-category scores for nurses were included, rather than those pooled with other 735 healthcare providers

ProQOL sub-category means/medians and SDs must include the raw, or 737 unnormalized because transformation to a t-score makes comparisons difficult due to the 738 mean (or median) being set at 50 with a SD of 10

• An additional approach to article selection included reviewing reference lists for pertinent 740 citations

Studies using a pre/post-intervention were included with the pre-intervention scores being 742 used