key: cord-0877649-gflv9pcc
authors: Dehon, Erin; Zachrison, Kori S.; Peltzer-Jones, Jennifer; Tabatabai, Ramin R.; Clair, Elizabeth; Puskarich, Michael A.; Ondeyka, Amy; Dixon-Gordon, Katherine; Walter, Lauren A.; Situ-LaCasse, Elaine H.; Fix, Megan L.
title: Sources of Distress and Coping Strategies Among Emergency Physicians During COVID-19
date: 2021-10-27
journal: West J Emerg Med
DOI: 10.5811/westjem.2021.9.53406
sha: 412c41cc9e67d1b8895363620b577574644b7158
doc_id: 877649
cord_uid: gflv9pcc

INTRODUCTION: The coronavirus disease 2019 (COVID-19) pandemic has been shown to increase levels of psychological distress among healthcare workers. Little is known, however, about specific positive and negative individual and organizational factors that affect the mental health of emergency physicians (EP) during COVID-19. Our objective was to assess these factors in a broad geographic sample of EPs in the United States. METHODS: We conducted an electronic, prospective, cross-sectional national survey of EPs from October 6–December 29, 2020. Measures assessed negative mental health outcomes (depression, anxiety, post-traumatic stress, and insomnia), positive work-related outcomes, and strategies used to cope with COVID-19. After preliminary analyses and internal reliability testing, we performed four separate three-stage hierarchical multiple regression analyses to examine individual and organizational predictive factors for psychological distress. RESULTS: Response rate was 50%, with 517 EPs completing the survey from 11 different sites. Overall, 85% of respondents reported negative psychological effects due to COVID-19. Participants reported feeling more stressed (31%), lonelier (26%), more anxious (25%), more irritable (24%) and sadder (17.5%). Prevalence of mental health conditions was 17% for depression, 13% for anxiety, 7.5% for post-traumatic stress disorder (PTSD), and 18% for insomnia. Regular exercise decreased from 69% to 56%, while daily alcohol use increased from 8% to 15%. Coping strategies of behavioral disengagement, self-blame, and venting were significant predictors of psychological distress, while humor and positive reframing were negatively associated with psychological distress. CONCLUSION: Emergency physicians have experienced high levels of psychological distress during the COVID-19 pandemic. Those using avoidant coping strategies were most likely to experience depression, anxiety, insomnia, and PTSD, while humor and positive reframing were effective coping strategies.

Prior to the coronavirus 2019 (COVID- 19) pandemic, physicians struggled with heightened levels of burnout, job dissatisfaction, depression, post-traumatic stress symptoms (PTSS), and suicidal ideation. 1, 2 Over the past year, emergency physicians (EP) were positioned as frontline caregivers for COVID-19, which further escalated challenges and pressure on the healthcare system and its workers.

Studies have shown that pandemics such as severe acute respiratory syndrome (SARS) 2003 and COVID-19 are associated with increased levels of healthcare worker psychological distress, including burnout, anxiety, depression, insomnia, and post-traumatic stress. [3] [4] [5] [6] [7] [8] [9] [10] During the early stages of COVID-19, distress was particularly high in healthcare workers without consistent access to personal protective equipment (PPE) 11 and those exposed to COVID-19 patients. 12 A systematic review of 59 internationally diverse studies revealed that psychological distress associated with COVID-19 is a global problem. 13 Studies of EPs, in particular, show increased levels of psychological distress in response to COVID-19. 10, 14, 15 One survey of over 400 EPs revealed increases in work stress, home anxiety, emotional exhaustion, and burnout. 14 Given that physicians are experiencing negative effects from the COVID-19 pandemic, it is critical to identify factors influencing physician stress for appropriate interventions to be designed. To date, there is limited data on which interventions have yielded the most success. Of the few published qualitative studies that have investigated potential contributors to physician anxiety, organizational factors such as access to PPE, exposure to COVID-19 at work, uncertainty of organizational support and lack of access to testing, childcare access and up-to-date information and communication were noted as main drivers. 16 Current EP-specific literature is limited. Most studies were performed outside the US or in limited geographical areas such as New York City. Additionally, many do not include measures of psychological distress with strong validity evidence. Furthermore, there is not, to our knowledge, any current data focusing on possible positive psychological reactions to COVID-19 or effective coping strategies. Finally, although some studies have looked at factors contributing to clinician stress, none have performed a comprehensive stepwise approach using an assessment of multiple contributory factors. Our aim in this study was to extend prior research by identifying both individual and organizational factors that place EPs at risk for psychological distress during COVID-19. Additionally, we sought to identify any positive effects related to COVID-19 and examine coping strategies used by EPs.

This was a prospective cross-sectional survey of EPs administered via email between October 6-December 29, 2020. Demographic and work-related data were collected from respondents. We assessed negative mental health outcomes, positive work-related outcomes, and strategies used to cope with COVID-19. All surveys were completed anonymously. This study was approved by the local institutional review board and is reported according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (Appendix 1). 17

Participants consisted of attending physicians who worked in an emergency department (ED) in the US during the COVID-19 pandemic. To recruit participants, we used a combination of convenience and purposive sampling strategies. A purposive sampling strategy was used to obtain a sample of EPs working in various US regions. We sent directed emails to a convenience sample of known colleagues who work at the identified hospitals asking them to function as survey champions by distributing the survey to all known EPs at their site who had worked in the ED during COVID-19. All participants received a $40 gift card for completing the survey.

Below 

An a priori power analysis using G*Power 29 software (University of Dusseldorf, Germany) indicated that the sample size needed to detect a medium effect was 194 based on an alpha of .05, power of .95, and 14 predictors. Preliminary analyses were conducted to test the assumptions for the regression analyses. We calculated response rate using American Association for Public Opinion Research (AAPOR) response rate 2 definition, which allows for the inclusion of both complete and partial surveys. 30 Non-response bias was evaluated by comparing the early and late participants' scores on mental health outcomes. 31 For all measures, we evaluated internal reliability using Cronbach's alpha. Construct validity was established by examining correlations with other theoretically related, psychological-outcome measures. Basic descriptive statistics and established cutoff scores, and diagnostic algorithms were used to examine the prevalence of PTSD, insomnia, depression, and anxiety among EPs. We used Mann-Whitney U tests to compare psychological outcomes among EPs by demographic and epidemic-related factors. To balance Type I and Type II error across the eight analyses for each of the three outcomes, we applied a Holm-Bonferroni correction. 32, 33 We performed four separate three-stage hierarchical multiple regression analyses to examine whether individual and organizational COVID-19 challenges related to COVID (eg, fear of COVID-19, PPE access) were predictive of psychological distress after accounting for demographic variables and living arrangements. In the final step, the unique contribution of coping strategies in predicting psychological distress was examined. The dependent variables were depression, anxiety, PTSS, and insomnia. In each regression, the predictor variables gender, age (> or < than 40 years old) and whether they were living alone (yes/no), living with children (yes/no), living with elderly individuals (yes/no), and isolated from family at any point during COVID-19 (yes/ no), and time of survey were analyzed in the first step. For the second step analysis, we added the predictor variables job stress, stigma, obsession with COVID-19, fear of COVID-19, and perceived adequacy of training, protection, and support. In the final step, we analyzed coping styles (approach and avoidant coping).

A total of 517 EPs representing 11 institutions across 11 different states were invited to complete the survey. The overall response rate using the AAPOR response rate 2 definition was 50%. This included 251 complete surveys and eight partially completed surveys (30-90% complete) . Surveys were completed between October-December 2020. Respondents were 63% male and 37% female. About half

Coping Strategies Among EPs During COVID-19

of the participants were aged 30-40. Ten participants (4%) had been infected with COVID-19-. The majority (95.5%) of participants reported having adequate PPE over the prior month. Additional characteristics of the study population are shown in Table 1 .

To assess for non-response bias we compared early respondents to initial non-respondents across all mental health outcomes, based on the assumption that late respondents were similar to non-respondents. 31 When comparing early respondents and initial non-respondents, we found no significant differences in levels of depression, anxiety, PTSS, or insomnia. The characteristics of early respondents and initial non-respondents are presented in Appendix 3. Furthermore, the proportion of female respondents in this sample (37%) is consistent with the proportion of academic EPs nationwide who are female (37%). 34

Supporting the validity of these measures, fear of COVID-19, obsession with COVID-19, perceived stigma, and job stress were linked to anxiety, depression, and insomnia in the expected directions. As predicted, obsession with COVID-19 and fear of COVID-19 showed slightly stronger associations with anxiety than with depression. A comprehensive correlation matrix can be found in Appendix 4. Internal consistency was acceptable across measures: job stress (α = .65); perceived stigma (α =.79); obsession with COVID-19 (α =.80); fear of COVID-19 (α = .87); and training, protection, and support (α =.87).

Based on a single-item measure of the overall impact of the pandemic, 85% of participants reported that COVID-19 has had some negative impact on their mental health. The Coping Strategies Among EPs During COVID-19 Dehon et al.

level of impact COVID-19 has had on EP mental health was described as large (12%); moderate (31%); and small (40%). Compared to how they felt pre-COVID-19, participants reported feeling more stressed (31%), lonelier (26%), more anxious (25%), more irritable (24%), and sadder (17.5%) ( Compared to the six months before COVID-19, participants who were exercising at least three days a week decreased slightly from 69% to 56%. The number of participants reporting daily alcohol use nearly doubled over the same period from 8% to 15%. Participants reporting some level of burnout increased from 16% to 41% (Table 2 ). Based on a single item, 14% of participants reported that their experiences working during COVID-19 had made them wish they had chosen a different specialty. Based on established cutoff scores and diagnostic algorithms, the prevalence of mental health conditions among the sample was 17% for depression, 13% for anxiety, 7.5% for PTSD, and 18% for insomnia.

Measures of organizational variables showed that increases in work-related stress (66%) and workload (63%) were prevalent. While most participants felt they had adequate training to work in the ED during COVID-19, only half felt appreciated and supported by their employer. Feeling stigmatized because of their work was also common (56%) ( Table 3) . Table 4 displays the association between pandemicrelated factors and psychological distress. Mann-Whitney U tests showed that EPs who reported isolating from family had significantly higher levels of depression (P < .001, effect size = .21); anxiety, (P = .003, effect size = .19); PTSS (P = .004, effect size = .18); and insomnia (P = .002, effect size = . 19 ). Anxiety levels were higher among EPs who reported lacking access to PPE (P = .006, effect size = .17) and staffing shortages (P = .003, effect size = .19) ( Table 3) . Experiences of PTSS were higher among EPs who reported ventilator shortages (P = .001 effect size = .21). Gender, age, and geographical region were not associated with levels of anxiety, depression, PTSS, or insomnia.

Positive effects of COVID-19 were also reported ( Table 5) . Overall, 84% were at least slightly satisfied (vs dissatisfied) with their current job. The majority of participants included feeling at least "a little" more appreciated by patients and society (65%), having a greater appreciation (74%) and enthusiasm (44%) for the job, and feeling an increased sense of togetherness among colleagues (87%). (Table 6) Next, we examined for characteristics independently associated with four mental health concerns. Models were

Neither Agree

Training protection and support Note: For the purpose of this table, we combined responses of "strongly disagree" "disagree," and "somewhat disagree" into one "Disagree" category. Responses of "strongly agree" "agree," and "somewhat agree" were combined into one "Agree" category. The full scale was used to calculate total subscale scores. ED, emergency department; PPE, personal protective equipment. .04 .0125 8 (4-12)

. 13 .008 

Coping

examined in a series to identify variation in individuals' mental health concerns that were attributable to basic individual factors (demographics, living arrangements), individual and organizational challenges related to COVID-19 (eg, fear of COVID, job stress, PPE access), and coping styles.

We found that 9% of variation in individuals' likelihood of depression was explained by basic individual factors, with isolation from family and later time of survey completion significantly associated with likelihood of depression symptoms. After accounting for basic individual characteristics, an additional 18% of variance in depression symptoms was explained by challenges related to COVID-19, with isolation from family, later time of survey completion, living alone, job stress, and obsession with COVID-19 significantly associated with likelihood of depression symptoms. After accounting for both basic individual factors and challenges related to COVID-19, coping behaviors predicted an additional 19% of the variance. The complete model explained 46% of the variance in depression. In the final model, living alone, isolating from family, job stress, and avoidant coping were significant predictors.

We found that 11% of variation in individuals' likelihood of anxiety was explained by basic individual factors, with female gender, living with children, later time of survey, and isolation from family significantly associated with anxiety symptoms. After accounting for basic individual characteristics, an additional 26% of variance in anxiety symptoms was explained by challenges related to COVID-19, with isolation from family, later time of survey completion, job stress, obsession with COVID-19, and fear of COVID-19 significantly predicting anxiety symptoms. After accounting for both basic individual factors and challenges related to COVID-19, coping behaviors predicted an additional 17% of the variance. The complete model explained 54% of the variance in depression. In the final model, female gender, living with children, later time of survey completion, job stress, and avoidant coping were significant predictors.

We found that 7% of the variance in PTSS was explained by basic individual factors, with those who isolated from family and who took the survey later in time reporting higher levels of PTSS. After accounting for basic individual characteristics, an additional 19% of variance in PTSS was explained by challenges related to COVID-19, with isolation from family, job stress, and obsession with COVID-19 significant predictors. After accounting for both basic individual factors and challenges related to COVID-19, the addition of coping behaviors predicted an additional 21% of the variance in PTSS. The overall regression model predicted 47% of the variance in .65 .05 7.5 (4) (5) (6) (7) (8) (9) (10) (11) (12) .87 .05 Coping Strategies Among EPs During COVID-19 Dehon et al.

PTSS. In the final model, isolation from family, job stress, and avoidant coping were significant predictors.

We found that 9% of the variance in insomnia symptoms was explained by basic individual factors with isolation from family and living with an elderly individual significantly predicting insomnia scores. After accounting for basic individual characteristics, an additional 9% of variance in insomnia symptoms was explained by challenges related to COVID-19, with age over 40, isolation from family, and obsession with COVID-19 significantly predicting insomnia scores. After accounting for both basic individual factors and challenges related to COVID-19, the addition of coping behaviors predicted an additional 6%. The complete model explained 24% of the variance in insomnia. In the final model, age over 40, isolating from family, and avoidant coping were significant predictors.

The most commonly used coping strategies among participants were acceptance, use of emotional support, planning, and self-distraction. We conducted four additional multiple regression analyses to examine which of the 14 specific coping strategies were associated with depression, anxiety, PTSS, and insomnia. Overall, results suggest that use of behavioral disengagement, self-blame, and venting were significant predictors of psychological distress. Humor and positive reframing were associated with lower levels of psychological distress. See Table 7 .

Despite recent attention to COVID-19's impact on the mental health of healthcare workers, this is the first nationally representative multisite study to examine its effect on US EPs. We found high levels of psychological distress due to the COVID-19 pandemic, but we also identified some positive effects from the pandemic. We also explored coping strategies that EPs used. Overall, 85% of our participants reported some negative impact on their mental health due to the pandemic. Compared to pre-pandemic levels, EPs were, on average, drinking alcohol more frequently, exercising less, spending less time with friends and family, and feeling more stressed, lonely, and anxious. This increase in negative effects is in line with many recent studies of healthcare workers in the time of COVID-19. [3] [4] [5] [6] [7] [8] [9] [10] [12] [13] [14] 35 Regarding specific outcomes, we found that a subset of individuals reported clinically elevated levels of insomnia (18%), depression (17%), anxiety (13%), and PTSD (7.5%). At first glance these prevalence rates may appear lower than rates of mental health concerns found in existing COVID-19-related studies. Several COVID-19-related studies have applied much lower cutoff scores (eg, PHQ-9 cutoff of 5 vs 10) and brief screening tools, 14, [35] [36] [37] which can lead to overestimates of prevalence rates. Rather than focusing on a narrow range of factors, this study adds to the literature by taking a comprehensive look at the impact of numerous individual (eg, demographic, fear of/obsession with COVID-19, coping strategies) and organizational (eg, practice setting, PPE, communication from leadership) factors as they relate to psychological distress. Throughout this pandemic, EPs have demonstrated resilience and the ability to adapt to a rapidly changing medical environment. Nonetheless, existing studies tend to focus on pathologizing EPs rather than highlighting factors that contribute to their resilience. This is not to suggest that the subset of EPs who are experiencing mental health concerns should be ignored. Rather, attention should also be focused on the vast majority of EPs who are not reporting high levels of distress despite the repeated day-to-day exposure to numerous stressors. In fact, compared to a sample of the general US adult population, EPs in the current study were reporting two times lower levels of anxiety and depression than the general population. 38 This was further echoed in the positive outcomes questions included in our survey in which 57% of respondents felt an increased sense of togetherness and cooperation among colleagues. Additionally, the majority of respondents reported feeling more appreciated by society.

A little less than half of the respondents reported having a greater appreciation for the value of his/her job, while onethird reported having greater job satisfaction as well as feeling more appreciated by patients.

In terms of individual variables, coping strategies were found to play a major role in predicting or protecting against negative impacts on mental health. Engaging in avoidance coping strategies, in particular, was found to be the strongest predictor of psychological distress across all of the individual, organizational, and pandemic-related factors examined. Avoidance coping strategies include denial, substance use, venting, behavioral disengagement, self-distraction, and selfblame. When looking at the coping strategies individually, behavioral disengagement emerged as a significant predictor of all four negative mental health outcomes. Venting and engaging in self-blame were also significant predictors of elevated depression, anxiety, and PTSS in our population. Of the "Approach'' coping strategies, use of "planning" as a coping response was significantly related to both depression and anxiety. Considering the uncertainty of COVID-19, it is understandable that a typically adaptive coping strategy (planning) was rendered ineffective during the outbreak. Positive reframing was also significantly negatively correlated with depression and anxiety in our population which helps explain why so many physicians reported experiencing positive outcomes from COVID-19. Table 7 . Multiple regression analyses for coping variables predicting mental health outcomes.

Note: Standardized beta coefficients are reported for comparability. *P < .05. **P < .01. SD, standard deviation; PTSS, post-traumatic stress symptoms.

Coping Strategies Among EPs During COVID-19 Dehon et al.

Humor, which is not considered an approach or avoidance strategy, was significantly negatively correlated with three of four of the main dependent variables (depression, anxiety, PTSS). Finding ways to incorporate humor in wellness interventions, staff meetings, education sessions, and even during shifts, may be a critical strategy not receiving enough formal attention. As a whole, these findings underscore the importance of offering individual-level interventions designed to promote the use of adaptive coping strategies and identifying at-risk colleagues who may be using maladaptive coping strategies.

Organizational factors also played a significant role in predicting physician distress. In prior studies addressing healthcare worker concerns during the COVID-19 pandemic, clinicians cited lack of PPE and isolation from family as major sources of anxiety. 16, 37 Our findings confirmed that both lack of access to PPE and isolation from family were positively correlated with increased levels of psychological distress including depression, anxiety, PTSS, and insomnia. Higher levels of psychological distress were more common among individuals who reported experiencing PPE, ventilator, and/or staffing shortages at any point in time over the course of the pandemic.

In terms of PPE, current access to PPE was not an issue for the vast majority of the participants during the time period of this study (October-December 2020) with 95.5% of respondents reporting that they had access to adequate PPE. Nonetheless, 54.8% of physicians reported that they did not have adequate access to PPE prior to the survey, and staffing shortages were also extremely common with 73.2% of respondents reporting shortages. Both limited access to PPE (at any point during the pandemic) and staffing shortages were associated with higher levels of psychological distress. In addition, physicians who were isolated from their families experienced higher levels of anxiety, depression, PTSS, and insomnia. Our findings emphasize the need for organizational support for those separated from their families via resources such as housing and/ or childcare. Increases in workload and increased job stress also had positive associations with anxiety, depression, and PTSS. Taken together, these findings highlight the importance of organizations supporting their physicians by ensuring adequate resources, staffing, and support during times of crisis.

Several limitations of this work deserve consideration. First, participants were a convenience sample of physicians from 11 hospitals who were identified based on known contacts at those sites; therefore, results may not be representative of the entire EP population. By limiting the number of participating programs (rather than distributing via listserves) we were able to maximize our response rate. Second, surveys were taken at a single point in time. Given the dynamic nature of the pandemic, physicians may have taken the survey before, during, or after a surge of patients. While we attempted to assess for this, these differences could have affected results. Similarly, longitudinal data were not available to assess how physicians responded to dynamic changes.

Third, the survey was targeted toward EPs at academic medical centers, and generalizability to community or rural sites is unknown. Fourth, while the hypotheses of the study were not explicit, a Hawthorne effect may have been present. Furthermore, despite the strength of the instruments used, it is possible other measures could have yielded different results. Finally, although many would consider our response rate acceptable and we found no evidence of non-response bias, there was still the potential for sampling bias.

Emergency physicians experienced high levels of psychological distress during the COVID-19 pandemic. Individuals reporting avoidant coping strategies were most likely to experience depression, anxiety, insomnia, and PTSD. In contrast, humor and positive reframing were effective coping strategies for physicians. Strategies focusing on positive work-related experiences during the pandemic such as increased feelings of societal value or appreciation and increased sense of camaraderie with colleagues may be of value. These findings highlight the importance of hospitals supporting physicians through offering interventions designed to promote the use of adaptive coping strategies.

This study was funded by the National Foundation of Emergency Medicine. Conflicts of Interest: By the WestJEM article submission agreement, all authors are required to disclose all affiliations, funding sources and financial or management relationships that could be perceived as potential sources of bias. No author has professional or financial relationships with any companies that are relevant to this study. This study was funded by the National Foundation of Emergency Medicine.

Committee on Systems Approaches to Improve Patient Care by Supporting Clinician Well-Being

Against Clinician Burnout: A Systems Approach to Professional Well-Being

Physician burnout: a global crisis

The psychological impact of the SARS epidemic on hospital employees in China: exposure, risk perception, and altruistic acceptance of risk

Prevalence of depression, anxiety, and insomnia among healthcare workers during the COVID-19 pandemic: a systematic review and meta-analysis

Factors associated with mental health outcomes among health care workers exposed to coronavirus disease 2019

Mental health outcomes among healthcare workers and the general population during the COVID-19 in Italy

Prevalence of depression, anxiety, distress and insomnia and related factors in healthcare workers during COVID-19 pandemic in Turkey

Psychological effects of COVID-19 on hospital staff: a national cross-sectional survey in mainland China

Psychological symptoms among frontline healthcare workers during COVID-19 outbreak in Wuhan

Canadian emergency physician psychological distress and burnout during the first 10 weeks of COVID-19: a mixed-methods study

Personal protective equipment and mental health symptoms among nurses during the COVID-19 pandemic

Exposure to COVID-19 patients increases physician trainee stress and burnout

The mental health impact of the Covid-19 pandemic on healthcare workers, and interventions to help them: a rapid systematic review

Academic Emergency medicine physicians' anxiety levels, stressors, and potential stress mitigation measures during the acceleration phase of the COVID-19 pandemic

Longitudinal prospective study of emergency medicine clinician wellness across ten academic and community hospitals during the initial surge of the COVID-19 pandemic

Understanding and addressing sources of anxiety among health care professionals during the COVID-19 pandemic

Strengthening the Reporting of Observational Studies In Epidemiology (STROBE) statement: guidelines for reporting observational studies

Longterm psychological and occupational effects of providing hospital healthcare during SARS outbreak

Fear of severe acute respiratory syndrome (SARS) among health care workers

How much "thinking" about COVID-19 is clinically dysfunctional?

You want to measure coping but your protocol's too long: consider the brief COPE

Using a single item to measure burnout in primary care staff: a psychometric evaluation

The Post-traumatic Growth Inventory: measuring the positive legacy of trauma

Transformational leadership and psychological well-being: the mediating role of meaningful work

The PHQ-9: validity of a brief depression severity measure

A brief measure for assessing generalized anxiety disorder: the GAD-7

The posttraumatic stress disorder checklist for DSM-5 (PCL-5): development and initial psychometric evaluation

Validation of the Insomnia Severity Index as an outcome measure for insomnia research

Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses

American Association for Public Opinion Research (AAPOR)

Poll-Survey-FAQ/Response-Rates-An-Overview

Improving response rates and evaluating nonresponse bias in surveys: AMEE Guide No. 102

A simple sequential rejective multiple test procedure

Holm-Bonferroni method: step by step" from StatisticsHowTo.com: elementary statistics for the rest of us!

Association of

The psychological impact of COVID-19 on hospital staff

Prevalence and factors associated with depression and anxiety among health care personnel in the United States during coronavirus disease 2019 (COVID-19) pandemic

Psychological distress, coping behaviors, and preferences for support among New York healthcare workers during the COVID-19 pandemic

Symptoms of anxiety or depressive disorder and use of mental health care among adults during the COVID-19 pandemic -United States