Utilization of Wellness Practices For Burnout and Stress During COVID-19 Among an Interdisciplinary Cohort of Emergency Healthcare Workers

cross-sectional 28-item

having clear communication, access to mental health support, confidence in infection control measures, and support from colleagues and leadership, all contribute to improved wellness in stressful work environments [37][38][39].A notable study addressed the organizational response to anxiety early in the COVID-19 pandemic included five requests from healthcare professionals to their organization: "hear me, protect me, prepare me, support me, and care for me" [40].
The COVID-19 pandemic presented a unique stressor, particularly within the healthcare community, that created a distinct opportunity to determine if any wellness interventions may reduce stress or burnout among EHCWs.Thus, our goals were to evaluate protective factors and wellness interventions that were utilized by EHCWs during the COVID-19 pandemic.Our primary objective was to determine the perceived benefit of wellness interventions by ECHWs so that recommendations for the most beneficial interventions could be prioritized.Additionally, we evaluated perceived stress and burnout among an interdisciplinary group of EHCWs during the COVID-19 pandemic in order to better understand the extent of the pandemic's impact on well-being.

METHODS
A cross-sectional survey of EHCWs at three large metropolitan hospital Emergency Departments (EDs) was used to measure participants' utilization and perceived benefit of individual and institutional wellness practices and resources, burnout, overall stress, and stress related to experiences from COVID-19.Eligible participants were EHCWs, including attending emergency physicians (139), resident emergency physicians (60), emergency medicine trained fellows (7), emergency medicine physician assistants and nurse practitioners (collectively referred to as advanced practice providers) (APPs, 125), emergency nurses (156), and ED staff (research and administrative staff, 96).Physician, APP, and ED staff lists were generated by the Department of Emergency Medicine, and a list of emergency nurses was generated from the hospital (N=583).The survey was initially sent via email on July 6, 2020 to physicians, APPs, and staff.Additional hospital approval was required for the nurses' survey resulting in a delay in distribution that also limited the nurse participants to only one of the three hospitals included.Surveys were emailed to the nurses on August 17, 2020, and the study link was closed for all participants on September 16, 2020.Any EHCW involved in the creation of this survey were excluded.
Participants completed the survey on an online and secure platform (Microsoft Forms).Most participants received seven reminders by email with a link to the survey, while nurses received four reminders due to the delay in their initial invitation.The study was approved by the academic institution's Internal Review Board and the associated hospital's Research Oversight Committee.No participation incentives were offered.
The survey consisted of 28 questions.Participants were asked to rate the usefulness and helpfulness of 11 different wellness activities and offerings on a five-point Likert scale, ranging from extremely helpful to extremely not helpful, and with the additional options of "service not available or offered to me" and "did not use."The following activities / offerings were assessed: free meals offered during clinical shifts, community-donated PPE, virtual yoga or meditation, counseling services, childcare resources, financial support, support while out due to COVID-19, flexible / reduced shift hours, other forms of community support, national / local retailer or restaurants offering free or discounted products, wellness committee buddy program, and zoom social meetings.
Additionally, the survey included the perceived stress scale (PSS) to measure general stress, a two-item inventory to measure burnout, and items to describe demographics, home and work life, and stress related to experiences of COVID-19.The authors generated subject matter questions specific to experiences of the COVID-19 pandemic and wellness interventions based on prior studies, experience in the ED, and discussions with colleagues.A statistician was consulted in the creation of the survey.The questionnaire was piloted for comprehension by six EHCWs and authors (APPs and physicians).
The PSS is a ten-item, validated questionnaire measuring the degree to which respondents find their lives unpredictable, uncontrollable, and overloaded [41].Higher stress scores are correlated with depressive symptoms and utilization of health services [41].Two items were used to measure emotional exhaustion and depersonalization-two widely recognized components of burnout.These items have been validated in medical professionals and found to provide meaningful information on burnout [42].The impact of COVID-19 on EHCWs home life and work life was measured using a five-point Likert scale, ranging from very unconcerned to very concerned.Participants rated concern regarding: contracting COVID-19, being overwhelmed at work, changes in work productivity, access to personal protective equipment (PPE), and the financial impact of COVID-19.
The response rate was calculated using the American Association for Public Opinion Research response rate definition which includes complete and incomplete responses divided by the sum of complete and incomplete responses, breakoffs, and all other non-respondents [43].Surveys were distributed only to eligible participants via their work email address; therefore, partial responses were included.Categorical variables were described using frequencies and percentages.Continuous / scale variables were described using medians and interquartile ranges.Comparisons between categorical variables were conducted using χ2 tests / Exact tests and comparisons of continuous / scale variables were conducted using Kruskal-Wallis and Mann-Whitney U tests.Odds ratios and 95% confidence intervals were computed using logistic regressions.Across the entire data set, 1.4% of the data were missing (range across variables: 0% -6.4%).Ten complete data sets (including partial responses) were imputed using fully conditional specification [44].All variables in the data set were used in the imputation models.

Wellness Activities/Interventions
Participants reported engaging in a variety of individual-level wellness activities, as summarized in Figure 1.Wellness offerings that were rated as extremely helpful included community-donated PPE (43.1%), meals offered during clinical shifts (36.9%), financial support such as government stimulus checks (29.2%), and retail / restaurant discounts or free items (24.2%) (Figure 1 ).

The Impact of Wellness Activities on Burnout
Table 2 presents the relationship between self-reported wellness activities and burnout.Most wellness activities were associated with lower reports of burnout but were not statistically significant.Exercise and engaging in hobbies were associated with lower burnout in unadjusted analysis, although this finding did not remain significant in the regression model.In the unadjusted analysis, respondents who spent more time with pets and respondents who saw a therapist / counselor regularly were more likely to report burnout.Following adjustment for: other wellness activities, age, gender, race, and EHCW role, these associations were still significant.

Burnout and Perceived Stress
Burnout rates and median PSS scores are reported in Table 3 (next page).Burnout varied significantly by gender and EHCW role.Females reported significantly higher rates of burnout than males across the cohort.The highest rates of burnout were noted in RNs, and this relationship was independent of gender.In subgroup analysis by EHCW role, female APPs had significantly higher burnout than did male APPs.A majority (58.5%) of respondents reported moderate to high or high stress overall, and half (50.0%) stated that they had felt nervous or stressed fairly often or often over the   last month.Nurses (both female and male) reported the highest median score on the PSS.In subgroup analysis by EHCW role, the median PSS for female residents was nearly significantly higher than median PSS for male residents.A majority (61.5%) also stated that they felt confident about their ability to handle their personal problems fairly often or often.

Stress related to COVID-19
The majority (70.8%) reported moderate to high or high stress resulting from COVID-19 with 28.8% very concerned about contracting the virus and 59.6% very concerned about a family member or friend contracting the virus.Other elements that resulted in high levels of concern included availability of PPE at work (28.8%) and the personal financial impact of the pandemic (31.2%) (Figure 2).
Many (75.0%) of those who tested positive for COVID-19 rated their overall stress as moderate-high or high, compared to 63.6% of those who tested negative and 52.5% of those who had never been tested (p = 0.04).

Work-life
Most participants reported no change in their work hours: 175 (67.3% of the total sample), including 80.0% of attendings, 73.6% of APPs, and 61.1% of registered nurses.Only 25 participants (9.6%) stated that their work hours decreased, 44% of whom were residents.About a quarter (23.1%) reported an increase in work hours, which was most common among staff (39.4%) and RNs (37.5%).Few APPs (20.3%), attending physicians (15.3%), and resident physicians (6.7%) reported increases in work hours.
Nearly a quarter (24.1%) of participants reported missing work due to testing, symptoms, or confirmed COVID-19.Almost one-half of participants had already been tested for COVID-19 (121, 46.5%) and of those tested, 101 reported negative results, and 20 (7.7% of all participants) reported testing positive.Of the 120 participants who were tested for COVID-19 antibodies in Spring 2020: 110 tested negative and 10 tested positive.Of all the EHCWs who responded, one NP, one resident, three attendings (4.2% of attendings), 11 registered nurses (15.3%), and four other staff (12.1%) had tested positive for acute COVID-19.

Home Life
A large minority (40%) of participants reported changing their living situation to avoid exposing others to the virus, and 27.3% reported difficulty with adult or childcare.A majority (77.7%) of participants reported feelings of loneliness during the pandemic.

DISCUSSION
EHCWs are exposed to a wide range of stressors and often more severe stressors than their colleagues in other departments [1][2][3][4][5].Our results suggest that burnout rates among EHCWs have not meaningfully deviated from the pre-pandemic nationally reported rates, despite the high rates of reported stress related to the COVID-19 pandemic.However, our reported rates of burnout are higher than other studies of EHCWs during COVID [25].
A critical and novel aim of this study was to determine the most commonly utilized and beneficial wellness activities by EHCWs.During the early pandemic response, an increased number of interventions were offered at the organizational level to provide support and improve well-being.These included meals offered during clinical shifts, community-donated PPE, virtual yoga or meditation, counseling services, childcare resources, a departmental wellness committee buddy program, and other social zoom meetings.Much of the recent literature on healthcare provider wellness has focused on shifting to an organizational / cultural approach and away from an emphasis on individual level initiatives and activities to mitigate burnout and improve well-being [45][46][47][48].A group of academic hospitals found success in utilizing wellness rounds, a wellness consult service for departments, and targeted wellness interventions for healthcare workers who were at risk of burnout [49].Others have identified a physical activity intervention to potentially improve well-being and reduce burnout during the early phase of the pandemic [50].In this study, participants reported high satisfaction with the helpfulness of several organization-level initiatives, including increasing PPE supplies from community donations, free on-shift meals, and financial support.We postulate that these interventions had high satisfaction as they offer not only tangible support but also a sense of recognition and gratitude for the work performed and risk of exposure experienced as EHCWs.
While institutional approaches to wellness remain very important, our results highlight the importance of meeting individual needs during periods of crisis, including protection against contracting the virus; this is documented in previous studies, as well [28,[37][38][39][40].Of note, some EHCWs surveyed received additional financial support from their employers during the pandemic, but the study question also referenced the spring 2020 United States (U.S.) government stimulus payments to individuals.Given the satisfaction with additional financial support, organizations may consider whether hazard pay would contribute to the wellness of EHCWs during future public health emergencies [51][52][53].
Participants also utilized individual level strategies to improve their well-being.The individual activities associated with lower reports of burnout were exercise, spending time with family / friends, spending time in nature, reflection activities, and engaging in hobbies or recreation.However, none were statistically significant in the adjusted analysis, highlighting the need for organizational level interventions and support.Since spending time with family / friends was associated with lower reports of burnout, it should be considered how the limitation in those activities during the pandemic may have contributed to decreased well-being.Nearly three-quarters of participants reported feeling lonely due to the social distancing guidelines, highlighting the importance of remaining connected with family, friends, and colleagues virtually or via socially-distanced events.
A scoping review of wellness strategies for healthcare worker wellness during the pandemic identified the importance of access to mental health resources and addressing the stigma of mental health in healthcare [54][55][56][57].Despite the significant stress and burnout reported, only about a quarter of respondents were seeking the assistance of therapists, which is consistent with prior studies on the attitudes toward and the utilization of mental health resources by HCWs [58][59][60][61][62].A 2020 survey by the American College of Emergency Physicians found that, despite higher levels of stress and burnout, 45% of emergency physicians did not feel comfortable seeking mental health treatment, and 73% felt there is a stigma of mental health treatment by others in their workplace [63].
A contributing factor to the stigma of the utilization of mental health resources in healthcare involves the questions that are asked on medical licensing applications.Work by Schroeder, et.al. demonstrated that 96% of allopathic medical licensing applications asked questions pertaining to the physical health, mental health, or substance use history of the applicant and that 69% of state medical licensing applications contained at least one item that was likely impermissible or impermissible based on the Americans with Disabilities Act (ADA) and appropriate case law [64].Despite the ADA, which has been in effect since 1990, there is ongoing concern that the presence of these types of questions on medical licensing applications may discourage and prevent physicians from seeking appropriate treatment due to fear of stigmatization, public disclosure, and impacts on licensure [64,65].There remains a call to action for state medical licensing boards to modify questions to address the issue of physician competence rather than medical history to protect applicant privacy and mitigate barriers to physicians seeking treatment for mental health concerns [64,66,67].
In our study, relatively few participants found counseling or mental health services to be helpful.In unadjusted analysis, we found that therapy was associated with a slightly increased risk of burnout, although we believe this association likely represents selection bias among those who choose to attend therapy.Alternatively, this finding may signify that participants with higher burnout were more likely to access mental health resources.Although there is less evidence to support a commensurate impact on burnout, several studies have found the benefit of mental health resources in reducing stress and anxiety [68][69][70].Given the elevated stress levels in our cohort, it remains important to eliminate the stigma of therapy and treatment for mental health concerns among EHCWs.
The COVID-19 pandemic resulted in an additional stress burden for EHCWs, including risks to personal and families' / friends' health [7,71].Participants who contracted COVID-19 were more likely to rate their overall stress as moderate-high or high.This may be due to the aforementioned stressors that are heightened when one has tested positive for COVID-19, including a higher likelihood of missing work, potentially infecting family / friends, having to change their living situation to avoid exposing others, and loneliness during isolation.Although not included in the survey, special consideration should be given to individuals who have suffered from chronic residual symptoms of COVID-19 (i.e., "long-haulers") [72].Institutions must have plans in place to support not only those with acute COVID-19 but also those who suffer from chronic residual symptoms of COVID-19 which may entail workplace modifications [72].
Interestingly, female APPs reported significantly higher burnout than did male APPs.This, along with the overall finding of higher burnout in females, may be due to the additional stressors of family care obligations.In future pandemics, organizations should consider stressors unique to individual employees and employee groups, which could include the need to address child and family care obligations.

Limitations
There are several limitations to this study.The primary limitation is a moderate response rate of 44.6%.While the reasons for the lower response rate are unknown, it is likely attributable to a combination of email fatigue, increased work stress, and the lack of incentives for study participation.It is also possible that those with higher rates of stress or burnout may have been more likely to respond to the survey.
Although the nurses' response rate of 46.2% was similar to the response rate across the entire cohort, it should be noted that nurses were emailed one month after the other participants due to a separate approval needed by the hospital's research committee.This resulted in less time for the nurses to complete the survey and may have contributed to their non-response rate.
This study was conducted in three academic hospitals in a large metropolitan area and our results may not be generalizable to non-academic, suburban, or rural hospitals.
The survey was administered from July to September 2020 which was several months after the first case of COVID-19 in the U.S..However, it did correlate with the second viral wave in the U.S., and one of the highest peaks in the state where the hospitals are located.[74,75].Thus, our survey may not have completely captured the study participants' stress and burnout levels from the initial pandemic response which included the acute shortages of PPE, lack of knowledge about the emerging disease, government-imposed shelter-in-place orders, and rapid changes in department operating procedures.Given the three-month response time for most participants, we may have captured data during fluctuations in participant well-being and burnout, which may have further influenced our results.
Although we utilized validated tools to measure stress and burnout, some of our data relies on participants' perceptions.We did not find a significant correlation between wellness activities and reduction in burnout as measured by the two-item inventory used.This is possibly due to the multifactorial nature of burnout which may limit the impact of any single wellness intervention.Because of this, our findings of the most helpful wellness interventions were primarily based on participant perception, which limits the reliability of our findings.There is the additional potential for recall bias (although this was mitigated by the use of validated instruments which targeted participants' current state).

CONCLUSION
The COVID-19 pandemic represented a significant public health crisis in the United States and was a significant contributor to the stress of EHCWs in 2020.Our study found that community and financial support were perceived as helpful to the well-being of EHCWs, and personal activities such as exercise were inversely correlated with burnout.The survey found higher rates of burnout in female participants and nurses, and higher levels of stress were reported in those who were diagnosed with acute COVID-19 and nurses.The higher rates of burnout and stress in nurses compared to other EHCWs highlight the need for additional wellness education, resources, and interventions specifically intended for the nursing profession.Additionally, organizations should consider ways to evaluate and support the unique needs of female EHCWs.While the need for systemic / organizational changes to support and improve wellness are of paramount importance, our study highlights that individual wellness practices should also be encouraged as they were perceived as helpful.We suggest future studies evaluate the benefit of individual wellness practices on stress mitigation and burnout reduction in EHCWs.
Funding Source: The author(s) received no specific funding for this work Conflicts of Interest: The author(s) have no conflict of interest to declare for this work.

*
Other group = Asian or Asian American; Middle Eastern or Arab American; Mixed; Other

Table 1 :
Demographic Characteristics of Participants

Table 2 :
Association between Wellness Activities and Burnout

Table 3 :
Report of burnout and median perceived stress scale according to healthcare role, gender and race Figure 2: Degree of Concerns Regarding COVID-19