key: cord-0732842-1sjdbo94 authors: Blanchard, Janice; Li, Yixuan; Bentley, Suzanne K.; Lall, Michelle D.; Messman, Anne M.; Liu, Yiju Teresa; Diercks, Deborah B.; Merritt‐Recchia, Rory; Sorge, Randy; Warchol, Jordan M.; Greene, Christopher; Griffith, James; Manfredi, Rita A.; McCarthy, Melissa title: In Their Own Words: Experiences of Emergency Health Care Workers During the COVID‐19 Pandemic date: 2022-03-24 journal: Acad Emerg Med DOI: 10.1111/acem.14490 sha: 9b08d5736d6f47c91d43a9a3a7ea5984c5cb9a70 doc_id: 732842 cord_uid: 1sjdbo94 BACKGROUND: During the COVID‐19 pandemic, a substantial number of emergency HCWs have screened positive for anxiety, depression, risk of post‐traumatic stress disorder and burnout. The purpose of this qualitative study is to describe the impact of COVID‐19 on emergency care providers’ health and well‐being using personal perspectives. We conducted in‐depth interviews with emergency medicine (EM) physicians, EM nurses and emergency medical service providers at ten collaborating sites across the United States between September 21, 2020, and October 26, 2020. METHODS: We developed a conceptual framework that described the relationship between the work environment and employee health. We used qualitative content analysis to evaluate our interview transcripts classified the domains, themes and subthemes that emerged from the transcribed interviews. RESULTS: We interviewed 32 emergency HCWs. They described difficult working conditions, such as constrained physical space, inadequate personnel protective equipment and care protocols that kept changing. Organizational leadership was largely viewed as unprepared, distant, and unsupportive of employees. Providers expressed high moral distress caused by ethically challenging situations, such as the perception of not being able to provide the normal standard of care and emotional support to patients and their families at all times, being responsible for too many sick patients, relying on inexperienced staff to treat infected patients, and caring for patients that put their own health and the health of their families at risk. Moral distress was commonly experienced by emergency HCWs, exacerbated by an unsupportive organizational environment. CONCLUSION: Future preparedness efforts should include mechanisms to support frontline health care workers when faced with ethical challenges in addition to an adverse working environment caused by a pandemic such as COVID‐19. interview transcripts classified the domains, themes and subthemes that emerged from the 139 transcribed interviews. 140 141 Results: We interviewed 32 emergency HCWs. They described difficult working conditions, 142 such as constrained physical space, inadequate personnel protective equipment and care 143 protocols that kept changing. Organizational leadership was largely viewed as unprepared, 144 distant, and unsupportive of employees. Providers expressed high moral distress caused by 145 ethically challenging situations, such as the perception of not being able to provide the normal 146 standard of care and emotional support to patients and their families at all times, being 147 responsible for too many sick patients, relying on inexperienced staff to treat infected patients, 148 and caring for patients that put their own health and the health of their families at risk. Moral 149 distress was commonly experienced by emergency HCWs, exacerbated by an unsupportive 150 organizational environment. 151 152 Conclusion: Future preparedness efforts should include mechanisms to support frontline health 153 care workers when faced with ethical challenges in addition to an adverse working environment 154 caused by a pandemic such as COVID-19 . 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 Introduction 172 173 The COVID-19 pandemic created a more hazardous and demanding work environment for 174 healthcare workers (HCWs). 1-3 Many frontline providers have faced high exposure to the virus, 175 increased working hours, and inadequate resources (e.g. personal protective equipment [PPE] , 176 personnel shortages, physical space constraints, insufficient infection control practices) for brief 177 or long periods of time. 1,2,4 These adverse working conditions have led to excess mortality [5] [6] [7] as 178 well as elevated rates of depression, anxiety and job-related stress, particularly among frontline 179 emergency HCWs. 8-12 180 181 Studies that focused on health outcomes of emergency HCWs during the pandemic reported that 182 a substantial proportion have screened positive for anxiety 10,13-16 , depression 13 17 risk of post-183 traumatic stress disorder 16 and burnout. 13,14,18-20 Most studies conducted to date relied on 184 standardized surveys that include scales and questions with uniform response options that 185 facilitate the quantification of outcomes and statistical comparison. 16, 18 Many qualitative studies 186 have focused on hospital-based providers without specific focus on EM HCWs and have 187 excluded emergency medical service providers. 21-23 188 189 The purpose of this study is to describe the impact of COVID-19 on emergency medicine (EM) 190 physicians, EM nurses and emergency medical services (EMS) providers' health and well-being 191 using a qualitative approach that allowed us to document participants ' reflections and 192 experiences in their own words. We relied on semi-structured interviews to identify important 193 themes and patterns that emerged from the personal narratives of the emergency HCWs. These 194 perspectives may not be obtained or well-understood with the use of quantitative survey data 195 alone . 196 197 Methods 198 199 Study Design. This investigation is the qualitative component of a mixed-methods study aimed 200 at evaluating the impact of COVID-19 on the health of emergency HCWs. This paper reports 201 the qualitative portion of the study. The project was a collaboration among ten academic sites. 202 We conducted in-depth interviews with frontline HCWs between September 21, 2020, and 203 October 26, 2020. The Institutional Review Board (IRB) at the institution of the principal 204 investigator (PI) served as the IRB of record and approved the study. We used the Standards for 205 Reporting Qualitative Research as a guide to reporting our results. 24 206 207 Theoretical Framework. We developed a theoretical framework based on two existing 208 occupational health models shown in Figure 1 . 25, 26 The first is an occupational safety 209 and health promotion model developed by Sorensen and colleagues that emphasizes the 210 centrality of conditions of work as determinants of employee health and safety. 25 This 211 model focuses on the degree of integration between an organization's policies and 212 practices related to occupational safety and health promotion. 213 214 Adverse working conditions cause job-related stress, which is not discussed in the Sorensen 215 model but is well-detailed in the psychosocial hazards of work model. 26 Job stress is the harmful 216 physical and emotional responses that occur when the job requirements do not match the 217 capabilities, resources or needs of the worker. The psychosocial hazards of work model proposes 218 that adverse working conditions such as work overload, unpredictable work hours, high 219 uncertainty and work-family conflicts negatively impact health directly or indirectly through the 220 experience of job-related stress. 26 221 222 Additionally, neither existing model includes potential moderators of the relationship between 223 workplace conditions and worker health nor acknowledges the role of the external environment 224 on the relationship between conditions of work and health. Our combined model posits that an 225 individual's coping behaviors and personal support can moderate the relationship between 226 working conditions and health. Our model recognizes that organizational characteristics and 227 conditions of work are strongly influenced by the policies, resources, knowledge, and attitudes at 228 the local, state, and national levels (see Figure 1 ). 229 230 Study Setting and Sample. We used purposive sampling to sample emergency providers 231 affiliated with 1 of 10 10 academic sites. Sites were chosen to capture diverse geographic 232 locations that varied in COVID-19 transmission rates, clinical caseloads, and geographic 233 location. An EM physician collaborator at each site identified one EM physician, one EM nurse 234 and one EMS provider for an in-depth interview. Our target sample size was one of each type of 235 emergency HCW from each site for a total of 30 participants across the ten sites. The goal was 236 to achieve a sample in which we would be able to capture rich content with sufficient saturation 237 of themes, without redundancy. 27,28 238 The EM collaborators provided the names and emails of potential participants to the co-principal 239 investigators (co-PIs, JB, MM) who contacted them directly. If we were unable to contact the 240 first HCW at the site, the site EM collaborator identified a second person for us to contact. 241 Participants consented verbally prior to the interview, and we also sent them a copy of the 242 consent form for their records. Collaborators were not involved in the interview process other 243 than recruitment. 244 245 We developed an interview protocol based on existing models describing the relationship 246 between job stress and health as well as studies that described stressors during prior pandemics, 247 particularly SARS and MERS 25, 26, [29] [30] [31] [32] . The interview guide is shown in the supplement. 248 249 We piloted our interview protocol with two emergency HCWs to test for content and clarity. We 250 finalized our protocol based on the pilot interviews and then the co-PIs (JB and MM) each 251 conducted half of the 1:1 interviews using Zoom. The interviews lasted approximately 45-60 252 minutes. All interviews were recorded and professionally transcribed. Each participant received 253 a $100 gift card upon completion. 254 255 The co-PIs started each interview by obtaining participants' demographic information. 256 Next, we asked participants open-ended questions about the main components of our 257 conceptual model including workplace conditions, stressors, mental and physical 258 wellness, and moderators of stress, both at the workplace and home. We used probing 259 questions such as "can you give me an example of this" or "please tell me more about 260 that" to obtain more detail about a theme of interest. Interviews continued until all pre-261 planned questions had been asked and data saturation was achieved. At the end of the 262 interview, we offered participants information about mental health resources as needed. 263 264 Data Analysis. We coded our transcript data using qualitative content analysis. 33-35 An initial a 265 priori codebook was developed deductively, based on our framework. 25, 26 This was followed by 266 an inductive process with a team consisting of a psychiatrist, health services researcher and 267 emergency medicine physician (JG, MM and JB) who reviewed and independently coded three 268 transcripts. We highlighted areas of text that mapped to our codebook and using open coding, 269 noted additional themes and subthemes that emerged. 35 270 271 The process was iterative, with team members meeting after each transcript to discuss themes 272 and subthemes, adjusting our codebook accordingly. Members of the coding team met to resolve 273 areas of convergence until agreement across all themes and subthemes had been achieved to 274 produce a final codebook. 275 276 After the codebook was finalized, two team members (JB and MM) coded additional transcripts 277 in NVivo, version 12.0. The two coders used a standardized approach to mark segments that 278 pertained to each theme and resolved disagreements by consensus after each transcript was 279 analyzed. 35 The process was repeated on successive transcripts until a high inter-rater reliability 280 score (Kappa >0.9) was achieved. After this, JB coded all 32 transcripts (including the two pilot 281 interviews) and MM reviewed and confirmed all codes. 282 283 We also used Nvivo to classify the domains, themes, and subthemes across emergency 284 HCW type. We summed the frequency of themes across individuals interviewed. We 285 selected illustrative quotes to demonstrate these themes. 286 287 288 Results 289 290 Of the 30 HCWs initially identified for participation, we completed an interview with 27. Two 291 who were initially recruited did not respond to emails and one declined to participate. We 292 replaced these three with a second HCW from the same site to reach our targeted sample size of 293 30. Those who completed interviews responded after an average of 1.5 attempts by email to 294 reach them, with the majority responding after the first attempt. Because our codebook did not 295 change after the pilot phase, we included our two pilot interviews. Table 1 displays the 296 demographic characteristics of our 32 participants. The mean age was 36 (range of 25 -58), and 297 slightly more than half of the sample was male (53%). Almost three-quarters of participants had 298 been working as emergency HCWs for at least five years. 299 300 301 Workplace conditions. Overall, participants described more difficult working conditions than 302 previously experienced because of the physical environment (n=26, 81%) and increased 303 workload (n=30, 94%) (see Table 2 ). With the exception of prolonged use of masks, most 304 participants reported that the availability of other types of PPE improved over the duration of the 305 pandemic. Space was especially challenging at peak periods; many participants stated that it was 306 difficult to physically separate patients under investigation for COVID-19 from patients who 307 presented with other medical problems. One attending stated, "Our entire department was just a 308 virus pit." In addition, prolonged boarding of patients in the ED was initially due to volume but 309 later persisted when some hospitals required COVID-19 test results prior to transfer to an 310 inpatient unit. 311 312 Almost all participants noted an increase in workload, not only because of patient volume but 313 also due to the cognitive processing involved in caring for patients with a disease that was not 314 well-understood. One attending physician said, "Everything changed on a day-to-day basis, and 315 then you were told you were doing things the right way. There was no right way." Protocols 316 frequently changed which required providers to spend excess time and effort making care 317 decisions. Residents and younger trainees reported lost learning opportunities in the workplace 318 setting. Their caseload was heavily skewed towards patients with COVID-19, giving them much 319 less exposure to typical medical emergencies. 320 321 The majority of participants had more positive comments than negative (n=23, 72%) about the 322 changes in the psychosocial culture of their working conditions (see Table 2 ). Even though 323 COVID-19 pervaded the work environment and led to low morale and work conflict at times, 324 many participants reflected positively on the closer bonds and collegiality they felt with their 325 fellow HCWs. One resident said, "It was teamwork. I felt like all of us were struggling, all of us 326 knew what needed to be done, and we just got right at it." 327 328 Organizational Characteristics. Many participants were frustrated with their organization's 329 response to COVID-19 at the highest leadership levels (n=25, 78%). They viewed leadership as 330 reactive and focused on output rather than providing staff with appropriate resources and 331 frequent communication (see Table 2 ). One resident stated, "It was honestly infuriating how 332 poorly we were prepared for this." Participants also felt that leadership was not transparent about 333 budgetary cuts and protocol changes. One nurse described that "on one horrible Saturday with 334 many sick, sick, patients… the CEO just came through and was pointing at people in this very 335 punitive way almost and he just says, 'Remember that you chose this'." In general, the EMS 336 providers were more positive about their employers' management of the COVID-19 pandemic 337 compared to the physicians and nurses. 338 339 Although few respondents reported that the organization provided them with hazard pay, lodging 340 or childcare assistance, almost all participants reported that there were mental health resources 341 available to them at the workplace to support them. Some reported being reluctant to take 342 advantage of the mental health resources because of time constraints, cost or concerns that it 343 would not remain confidential and could negatively affect their career. Consequently, only one 344 participant reported taking advantage of mental health resources at their workplace. Instead, 345 participants told us they relied on peer and/or family support to help them alleviate work stress. 346 347 External Environment. The impact of the external environment on the health and well-being of 348 emergency HCWs was mixed. More than half of the participants appreciated the community 349 support they received through the provision of food, housing, and public displays of gratitude 350 (n=20, 63%). On the other hand, some emergency HCWs experienced contagion stigma. For 351 example, one Asian American provider noted contagion stigma in the form of racial bias, citing 352 instances in which patients refused treatment because of stereotypical perceptions of disease risk. 353 Participants voiced frustration with social media and news stories that presented misinformation 354 and described how stressful it was to be working so hard to care of people who were not 355 attempting to adhere to public health guidelines (N=21, 66% 3 ). We listened to many instances of moral distress, which is the 361 psychological stress that individuals experience when they make decisions or perform tasks that 362 deviate from their personal values and sense of identity due to institutional or other constraints. 363 38, 39 The pandemic took a toll on the health and well-being of emergency HCWs in many different 388 ways (see Table 3 ). Many participants noted feelings of anxiety related to the changing 389 information about the disease and concern for family transmission. They described feeling sad 390 when telling families they could not see their loved one and hopeless when patients died alone. 391 Participants reported symptoms of depression, emotional exhaustion, and burnout from not 392 knowing how best to care for patients, the unpredictability of patients' trajectories and the many 393 deaths that occurred to young and old alike. One resident stated, "I felt just so drained." 394 395 Moderators of Stress. Home support helped lessen stress, especially when other family members 396 were in the medical field (n=16, 50%). However, broader support outside of the immediate 397 family was not a given. One resident said, "My family doesn't really understand, and they 398 would just freak out." Many participants described family members that did not live with them 399 who were reluctant or refused to see them because they feared contracting the virus from them. 400 Participants who lived alone or far from family members described isolation, which was 401 exacerbated by the restriction of normal social outlets during the pandemic (n=19, 60%). 402 403 Participants tended to report more positive coping mechanisms (n=29, 91%) than negative ones 404 (n=13, 41%). Many noted spending time with family, exercising, hobbies, meditating and getting 405 pets as sources of positive coping. Some also noted various adaptive coping strategies they used 406 during the pandemic such as coming up with creative ways to address limitations at work and 407 home. While some did discuss negative coping strategies, such as increased alcohol use and 408 overeating, these were generally not described as problematic or being associated with longtime 409 use. Discussion 412 413 This is the first qualitative investigation of the impact of the work environment on the health and 414 well-being that included EM providers across multiple roles (EM physicians, EM nurses and 415 EMS providers) and at diverse sites providers during COVID-19. Emergency HCWs experienced 416 many instances of moral distress, anxiety, depression, and exhaustion during the pandemic. In a 417 work environment that was more hazardous and cognitively taxing as compared to baseline, 418 many felt that their organization did not adequately protect and support employees' health, well-419 being, and safety. They commonly criticized organizational leadership, citing inconsistent 420 communication, lack of presence and insufficient guidance on handling the problems that arose. 421 Disinformation by social media and non-adherence to public health guidelines exacerbated their 422 job stress. Although mental health resources were usually available at work, emergency HCWs 423 relied largely on their immediate colleagues and family, exercise and hobbies for emotional 424 support and stress relief. 425 426 Consistent with previous studies of COVID-19, our participants reported that COVID-19 caused 427 them to feel more anxiety, depression, and burnout. 14, 16, 18, 19 Studies have described the anxiety 428 emergency HCWs felt when faced with the dilemma of spreading the virus to their loved ones at 429 home. 9, [8] [9] [10] [11] [12] [36] [37] One of the unexpected findings to emerge from this qualitative investigation was 430 how common emergency HCWs experienced moral distress. As a result, we revised our 431 theoretical framework to add moral distress as one of the mental health outcomes (see Figure 1 ). 432 Since the pandemic, there has been greater focus on the prevalence of moral distress and moral 433 injury in HCWs. 36,38-44 A recent study described the common experiences of moral injury among 434 HCWs across disciplines during the pandemic, however it did not focus specifically on 435 emergency providers and did not include the EMS population. 23 The results of our study must be interpreted in the context of the following limitations. 460 Participants were interviewed at different points in time in relationship to the peak of the 461 pandemic at their respective sites and there may have been some recall bias. We interviewed 462 individuals recruited by the site PI and therefore there may have been some selection bias. We 463 may not have captured a representative sample of EM physicians, EM nurses and EMS 464 providers, and may have interviewed individuals with fewer or more grievances than the general 465 population of frontline HCWs. Individuals may also have been less likely to report negative 466 behaviors during the interview. We interviewed providers from select institutions. Although 467 some physicians and nurses worked at community-based hospitals, these sites were affiliated 468 with academic centers, and therefore may have caused other potential biases. We also collected 469 limited data so do not present detailed demographics, such as marital status or caretaker 470 responsibilities, which may have affected perspectives. 471 472 We did not use scales to measure any of our mental health outcomes, but instead relied on 473 providers' personal narratives to identify common themes using qualitative research methods. 474 Therefore, we describe participants' feelings and symptoms, not experiences of disease. We also 475 did not measure or describe the general public health climate that may have impacted provider 476 opinions. Our conceptual model has not been previously validated, though it is based on two 477 existing, validated worker health models. In addition, we restricted interviews to EM physicians, 478 EM nurses and EMS providers. We did not interview other EM staff that may have been 479 similarly impacted by coping mechanisms during the pandemic, such as advanced practice 480 providers, security and environmental services staff. Finally, the generalizability may be limited, 481 since we do not know whether the race or gender distribution of our sample represents the 482 overall population of providers at our sites. Generalizability may have also been limited since 483 our study captured reflections during the early part of the pandemic. 484 485 486 Conclusion 487 488 In conclusion, this qualitative investigation revealed substantial moral distress among frontline 489 emergency care providers that has been largely under reported during COVID-19. COVID-19 490 revealed underlying deficiencies in preparedness that will need to be addressed if we are to 491 maintain a healthy and productive healthcare workforce during future pandemics . 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 References 537 At least 7,000 healthcare workers have died globally 12 Months of Trauma: More than 3,600 US Health Workers Died in Covid's 554 WHO estimate: 115,000 health workers have died from Covid-19, as calls for 557 vaccine access grow From Recession to Depression? Prevalence and 561 Correlates of Depression, Anxiety, Traumatic Stress and Burnout in Healthcare Workers 562 during the COVID-19 Pandemic in Greece: A Multi-Center, Cross-Sectional Study Hospital-Based Health Care Worker Perceptions of 565 Personal Risk Related to COVID-19 Factors Associated With Mental Health Outcomes Among 568 Health Care Workers Exposed to Coronavirus Disease Mental Health Outcomes Among Frontline and Second-571 Line Health Care Workers During the Coronavirus Disease 2019 (COVID-19) Pandemic 572 in Italy The mental health of medical workers in Wuhan, China 574 dealing with the 2019 novel coronavirus Pandemic-related mental health risk among front 577 line personnel Longitudinal Prospective Study of Emergency 579 Medicine Provider Wellness Across Ten Academic and Community Hospitals During the 580 Initial Surge of the COVID-19 Pandemic The experiences of health-care providers during the 582 COVID-19 crisis in China: a qualitative study Symptoms of Anxiety, Burnout, and PTSD 585 and the Mitigation Effect of Serologic Testing in Emergency Department Personnel 586 During the COVID-19 Pandemic Psychological Impact of the COVID-19 Outbreak on 589 Nurses in China: A Nationwide Survey During the Outbreak Anxiety Levels, Stressors, and Potential Stress Mitigation Measures During 593 the Acceleration Phase of the COVID-19 Pandemic Burnout amongst emergency healthcare workers 596 during the COVID-19 pandemic: A multi-center study Emergency physician stressors, concerns, 599 and behavioral changes during COVID-19: A longitudinal study Trainee doctors' experiences of learning and well-602 being while working in intensive care during the COVID-19 pandemic: a qualitative 603 study using appreciative inquiry COVID-19 confessions: a qualitative 606 exploration of healthcare workers experiences of working with COVID-19 Morally Injurious Experiences 609 and Emotions of Health Care Professionals During the COVID-19 Pandemic Before 610 Vaccine Availability Standards for reporting 613 qualitative research: a synthesis of recommendations Integrating worksite health protection and 616 health promotion: A conceptual model for intervention and research Health Impact of Psychosocial Hazards of Work: An Overview Characterising and justifying sample size 621 sufficiency in interview-based studies: systematic analysis of qualitative health research 622 over a 15-year period Qualitative methods for health research Factors associated with the psychological 626 impact of severe acute respiratory syndrome on nurses and other hospital workers in 627 Long-term psychological and 629 occupational effects of providing hospital healthcare during SARS outbreak. Emerg 630 SARS/MERS/SARS-CoV-2 Outbreaks and Burnout Syndrome among Healthcare An Umbrella Systematic Review Occurrence, prevention, 636 and management of the psychological effects of emerging virus outbreaks on healthcare 637 workers: rapid review and meta-analysis Three approaches to qualitative content analysis A standardized approach to qualitative content 641 analysis of focus group discussions from different countries The coding manual for qualitative researchers. 3 rd ed. Los Angeles ; London: 644 Sage Addressing caregiver moral distress 646 during the COVID-19 pandemic Psychological impact of COVID-19 outbreak on 648 frontline nurses: A cross-sectional survey study Clarifying the Language of Clinician Distress Reframing Clinician Distress: Moral Injury Not Burnout Assessment of the Association of Leadership 655 Behaviors of Supervising Physicians With Personal-Organizational Values Alignment 656 Among Staff Physicians Understanding and Addressing Sources of Anxiety 659 Among Health Care Professionals During the COVID-19 Pandemic Assessing civility at an academic 662 health science center: Implications for employee satisfaction and well-being Moral distress in frontline healthcare 665 workers in the initial epicenter of the COVID-19 pandemic in the United States: 666 Relationship to PTSD symptoms, burnout, and psychosocial functioning. Depress 667 Anxiety Addressing Moral Injury in Emergency 669 Covid-19: Eight in 10 doctors have experienced moral distress during 671 pandemic, BMA survey finds Holding the frontline: a cross-sectional survey of 676 emergency department staff well-being and psychological distress in the course of the 677 COVID-19 outbreak Facing Covid-19 in Italy -Ethics, Logistics, and Therapeutics on the 680 Epidemic's Front Line Protecting the front line: a cross-sectional survey 683 analysis of the occupational factors contributing to healthcare workers' infection and 684 psychological distress during the COVID-19 pandemic in the USA 688 breen-health-care-provider-protection 689 50. Council on Foreign Relations. Improving Pandemic Preparedness: Lessons from COVID-690 19 As an Asian American, the amount of discrimination and hate crimes… there was times that either they refused aid and they would want to request another ambulance. (