Balancing Provider Stress and Resilience in the Time of COVID Balancing Provider Stress and Resilience in the Time of COVID

Purpose: The purpose of this evidence-based practice project is to improve secondary traumatic stress (STS), compassion satisfaction, and burnout amongst providers— physicians, residents, and nurse practitioners—within the acute psychiatry units of the La Jolla Veterans Health Administration (VHA) through a one-time educational training session. Background: Mental health providers at the VHA acute psychiatry units experience one of the highest risks for the development of STS, CF, and burnout amongst all professions due to a number of individual and institutional factors. STS is characterized by secondhand traumatization with symptoms similar to that of Post-Traumatic Stress Disorder—difficulty sleeping, mood changes, upsetting images appearing in one’s mind, etc.—from repeated vicarious exposures via direct patient. CF results in a mental and/or physical detachment and a reduced ability to remain empathetic following repeated or prolonged interactions demanding high amounts of empathetic engagement. Both STS and CF significantly contribute to the development of burnout syndrome, which exacts a physical and psychological toll on the individual, is associated with poorer patient outcomes, and is costly to an organization. Methods: The intervention consists of a one-time educational class delivered to mental health providers designed to increase resilience against STS, CF, and burnout. The Professional Quality of Life (ProQOL), a 30-item Likert scale questionnaire with subscales for compassion satisfaction, burnout, and CF, was administered pre-and 30-day post-intervention. Results were gathered and entered into Intellectus Statistics online


List of Tables
computer software for analysis, and a t-test was conducted between each pre-and postintervention ProQOL sub-scale.
Results: Prior to the intervention, the group (n=13) had mean scores of compassion satisfaction that fell within the moderate levels, burnout scores within the upper end of low levels, and STS scores within low levels. Results from post-test scores (n=10) indicate very mild improvements in STS scores with no statistically significant changes in any sub-scale.
Implications: This project underscores the importance of holistic health and the benefits of self-care and is congruent with past studies showing STS, CF, and burnout are challenging problems to address. Though this intervention failed to produce significant changes to ProQOL scores amongst the VA's acute psychiatric provider population, burnout remains a pervasive issue, and further individual and/or institutional interventions are warranted.

Clinical Problem
Secondary traumatic stress (STS), compassion fatigue (CF), and burnout are three different but inter-related conditions possessing a constellation of symptoms affecting many healthcare providers; however, due to the often-intense nature of interactions with the patients receiving their care, mental health providers are particularly prone to developing these symptoms (Brady et al., 2012). There is growing concern that mental health providers at the Veterans Health Administration (VHA)-the largest integrated mental health care system in the United States-are at higher risk of burnout, and the evidence-based treatment strategies for post-traumatic stress disorder (PTSD) often include persistent focus on a patient's traumatic material which results in VHA workers experiencing frequent, sustained, vicarious trauma. (Garcia et al., 2015). While studies have shown burnout to be widespread throughout the mental health field, a 2016 survey by Garcia et al. found that 77% of VHA mental health providers considered emotional exhaustion, a key characteristic of burnout, to have impacted their quality of care. STS is characterized by vicarious traumatization with symptoms similar to that of PTSD-difficulty sleeping, mood changes, upsetting images appearing in one's mind, avoiding reminders of the event, etc.-from repeated secondhand exposures via direct patient care, while CF results in a mental and/or physical detachment and a reduced ability to remain empathetic following repeated or prolonged interactions demanding high amounts of empathetic engagement-both significantly contribute to the development of burnout syndrome (Wood et al., 2017). Stamm (2010) further elaborates: "People can experience negative effects of secondary exposure without developing a psychological disorder such as PTSD. Compassion Fatigue is not a diagnosis. It is possible that Compassion Fatigue is a descriptive term and that a person struggling with Compassion Fatigue also has a psychological disorder. For example, people who suffer with burnout may also have a diagnosable level of depression. Similarly, people may have a diagnosable level of PTSD or some other mental, emotional or physical disorder that is likely linked to their experience of compassion fatigue." As opposed to STS's typically acute onset and relation to one particular event, burnout syndrome-now recognized as a classifiable illness by the World Health Organization (World Health Organization, 2015)-has a gradual onset, culminating in a potentially chronic condition typically characterized by three key components: emotional exhaustion, depersonalization/cynicism, and reduced personal accomplishment/efficacy (Dreisen et al., 2018). When compared to other specialties, mental health workers face an increased risk of developing burnout due to unique challenges aiding in its development including stigma of the profession, demanding therapeutic interactions, actual or threatened violence from patients, and patient suicide (Eliacin et al., 2018).
Factors contributing to the development of burnout syndrome have been identified across three domains: personal, job, and organizational characteristics (Eliacin et al., 2018). Of the organizational characteristics, a review of the literature suggests high acuity, poor staff-to-patient ratios, heavy workloads, and low employee retention rates affect both burnout and quality of car while frequent clinical supervision, a sense of fair treatment, acknowledgment for doing well, and perceived autonomy are identified as protective factors (Humphries et al., 2014). A systematic review of determinants found those working in inpatient settings reported a lower sense of autonomy compared to community and specialist teams, who felt more autonomous with a greater sense of personal accomplishment (O'Connor et al., 2018). The same review also found a consistent correlation between rising age and increased risk of depersonalization but also increased feelings of personal accomplishment; furthermore, job instability and staffing shortages were associated with increased burnout rates (O'Connor et al., 2018). One study found that although hearing clients discuss traumatic content bothered over half of providers, it was patient characteristics such as difficult personality disorders and malingering that were more associated with burnout (Garcia et al., 2016). Another important cause of burnout is moral injury. Defined as "the damage done to one's conscience or moral compass when that person perpetrates, witnesses, or fails to prevent acts that transgress one's moral beliefs, values, or ethical codes of conduct" (Houtrow, 2020), moral injury has worsened during the coronavirus disease 2019 (COVID-19) due to limited resources and a considerable elevation in the mental health toll. In light of this, healthcare workers have faced being forced to alter treatment decisions due to circumstances beyond their control-whether it be a global pandemic or perceived institutional constraints (Houtrow, 2020). Furthermore, a meta-analytic study showed younger age to be a significant factor in emotional exhaustion and depersonalization in nurses, though no significant association was found between age and reduced personal accomplishment; additionally, marriage was found to be a protective factor for emotional exhaustion, and unmarried male nurses were shown to be most vulnerable in this regard (Gómez-Urquiza et al., 2017).
Considerable evidence shows burnout accounts for a damaging toll on both the physiological and psychological well-being of an individual, the functioning of their healthcare team, and is associated with decreased productivity, absenteeism, and compromised quality of care to the patients (O'Connor et al., 2018). The effects of repeated vicarious traumatization and the development of burnout syndrome, when combined with those factors detailed above, lend itself to undermining a mental health provider's sense of purpose and compassion, and have been shown to create higher rates of substance use, depression, and suicidality (Tsai et al., 2020). Unfortunately, physicians and nurses have a higher suicide rate than the general population, and although these deaths' relation to burnout may be unknown, recent research indicates the rates of depression in nurses and physicians to range from an alarming 25% to 43% (Melnyk, 2020).
The physiological response to stress is well documented and entails increased brain arousal initiated by neurotransmitters, catecholamines, and hormones such as cortisol which enact a sequence of bodily changes, preparing the body for its adaptive 'fight or flight' response (Winwood et al., 2006). When this adaptive stress response, which is designed for only intermittent use, is continually activated, the enduring presence of stress hormones creates the potential for a myriad of negative downstream effects including headaches, muscle tension, anger, impaired memory, decreased attention, anxiety, depression, reduced immune system efficiency, obesity, stroke and several other cardiometabolic effects (Winwood et al., 2006). A study examining burnout profiles amongst all VHA employees revealed a range of psychological and physical health problems reported by those with higher levels of burnout (Schult et al., 2018). The result of physical and psychological changes on the body may also lead to inter-personal conflicts and maladaptive coping mechanisms such as drug or alcohol use, which has a high concomitance with burnout syndrome (Mealer, 2016).
In addition to the drastic personal toll of burnout, this syndrome also has dire consequences for the institution in which that individual works. Regardless of the reason or method of leaving the profession, lost revenue per physician who leaves a practice is estimated to range from $500,000 to $1,000,000 (Melnyk, 2020), coupled with the $160,000 to $1,000,000 cost of replacing the physician-dependent upon specialty and experience-and these estimates do not include intangible losses such as team disruption and patient inconvenience (Stehman, 2020). Although data on the specific cost of burnout to the VHA is unknown, nationally, research estimates approximately $4.6 billion in costs related to physician turnover and decreased clinical hours each year within the United States (Han et al., 2019). The loss of a single provider from the profession is assessed to impact approximately 2,300 patients (Han et al., 2019), and with a shortage of mental health providers and a notable increase in psychiatric utilization during the COVID-19 pandemic (Bowman et al., 2021), every provider lost is of substantial consequence. STS, CF, and burnout not only have implications for the provider and employer but also can impact the client by degrading the therapeutic alliance, affecting communication, and increasing the likelihood of medical errors which may result in low consumer satisfaction and poorer patient outcomes including higher rates of hospitalization (Tsai et al., 2020). In summary, STS, CF, and burnout may enact significantly negative effects in the provider, the institution, and the patient; furthermore, those working in the acute psychiatry units within the VHA are subject to a higher risk of developing these negative effects.

Purpose
The purpose of this evidence-based practice (EBP) project is to improve STS, compassion satisfaction, and burnout amongst providers in the acute psychiatry units within the La Jolla VHA through a one-time educational training session as measured by the Professional Quality of Life (ProQOL) scale administered before and one month following the training. The acute psychiatry units consist of the psychiatric emergency clinic and inpatient units, and are staffed by physicians, residents, and nurse practitioners for whom the class will be administered. The resiliency training given to providers entailed descriptions of CF, STS, burnout, their effects and factors, and methods to improve resiliency including the introduction of breathing exercises, Metta/compassion meditation, and the Provider Resilience Mobile Application (PRMA).

Literature Review
To identify the evidence-based solutions for improving resiliency and mitigating STS, CF, and burnout, a review of the literature was conducted utilizing the electronic databases of CINAHL and PubMed. Key search terms included combinations of the  Reduce administrative burden 4. Enable technology solutions 5. Provide support to clinicians and learners, and 6. Invest in research on clinical professional well-being" (Melnyk, 2020). Typical individually based interventions seek to either increase mindfulness or to reduce negative arousal that characterizes stress in general though the use of relaxation techniques, promotion of a healthy lifestyle (diet and exercise), and cognitive behavioral techniques (Villani et a., 2013). Interestingly, a recent study found an inverse relationship between levels of resiliency and burnout and higher levels of resilience amongst all physicians when compared to the general working public; however, burnout rates remained substantial even among the most resilient physicians (West, et al., 2020). These findings are congruent with a 2017 systematic review and meta-analysis by Panagioti which found individual interventions for physician burnout to only produce small benefits but may be further enhanced by institutional-directed approaches.
A meta-analysis of burnout research from the last 35 years found that although the average effect of all interventions were relatively small, person-centered interventions are shown to be more effective than organization-centered interventions at reducing emotional exhaustion while organizational interventions such as clinical supervision, support groups, job restructuring, and team communication building had no significant effect on burnout (Dreison et al, 2018). Among organizational interventions, job training and education were found more effective than other subtypes at reducing overall burnout and feelings of personal accomplishment, though neither had significant effect on emotional exhaustion (Dreison et al., 2018).
A systematic review analyzing interventions on reducing burnout in physicians and nurses found workplace appreciation to produce a profound effect in increasing performance and decreasing depression and burnout, while team-based and coping interventions had no significant effect on healthcare worker burnout (Aryankhesal et al., 2019). Additionally, internet-based interventions were shown to improve mental health and reduce symptoms of depression, and interventions such as yoga, meditation, and mindfulness increased self-care and reduced emotional exhaustion (Aryankhesal et al., 2019).
The findings from the systematic review and meta analyses above suggest different modalities of interventions may be uniquely suited to address different burnout characteristics-emotional exhaustion, cynicism, or reduced personal accomplishment; however, according to a systematic review and meta-analysis conducted by West et al. (2016), the evidence remains unclear as to which interventions most improve resilience, though the most commonly studied methods involved mindfulness, stress management, and small group discussions, whose results show promise as effective measures in reducing burnout scores.
A meta-analysis examining mindfulness-based stress reduction (MBSR)-which typically includes methods such as controlled breathing, body scanning/progressive muscle relaxation, and meditation-found MBSR to be "moderately effective in reducing stress, depression, anxiety and distress and in ameliorating the quality of life of healthy individuals; however, more research is warranted to identify the most effective elements of MBSR" (Khoury et al., 2015). While this evaluation of studies was not specific to mental health providers, the authors noted that healthcare professionals were found to be among the populations who benefited most from MBSR (Khoury et al., 2015). A systematic review examining the effect of MBSR-based interventions on empathy and emotional competencies specifically in healthcare providers found it enhanced one's ability to regulate emotion with improvements in stress, psychological health, and empathy (Lamonthe et al., 2016). One method of MBSR includes the practice of deep breathing, which has been shown to increase positive effects on both psychological and physiological stress through the reduced activation of the limbic and sympathetic nervous systems; and in turn, these improvements positively affect cortisol levels, anxiety, blood pressure, and other chronic diseases influenced by stress (Hopper et al., 2019).
Another mechanism in which MBSR cultivates beneficial effect is through the nurturing of self-compassion. Research shows:

Description of EBP Project, Facilitators, and Barriers
Past studies have shown support groups and ongoing training interventions suffered from poor attendance due to ongoing staffing shortages common in acute psychiatry units, and ironically, the stress, burnout, and high turnover these organizationdirected interventions were meant to address, played a part in undermining the intervention (Gilbody et al., 2006). That being said, however, after inquiring within the La Jolla VHA, it was ascertained there was neither follow-up nor ongoing training in regard to CF, STS, or burnout being conducted after the initial new hire onboarding process, and supported by the evidence above, a clear need exists for such an intervention. Considering the evidence, and with the resources and timeframe at hand, it was deemed most appropriate to conduct a single educational class addressing the problem of CF, STS, and burnout scheduled during a regular administrative/continuing education period to lessen the risk of being perceived a burden. Although stress reduction techniques such as breathing exercises and progressive muscle relaxation are mentioned during the presentation, given the presumption most psychiatric providers already know-and likely teach their patients-these methods, more time was spent introducing LKM and the PRMA. follow-up reminders, ten responses were recorded; down from 13 initially. Reasons hypothesized for this loss to follow-up include not checking/seeing the emails, a preoccupation with patient care or other tasks, or a failure to remember.

Project Development and Timeline
Foremost, support was attained from the DNP student's faculty advisor, cosection chiefs, and other stakeholders at the La Jolla VHA acute psychiatry units.
Secondly, following a description of the EBP project and assurance of anonymity of ProQOL responses from the mental health providers, permission was granted by the labor representative on behalf of National Nurses United Veterans Affairs Division.
Subsequently, IRB approval was attained from both the La Jolla VHA and University of San Diego. The ProQOL was made available to members of the acute psychiatry units prior to the resilience training. In lieu of an in-person presentation, the resiliency training was conducted utilizing the Webex video conferencing platform in accordance with the La Jolla VHA's COVID-19 safety precautions, and a 30-minute PowerPoint presentation was given. Following 30 days from the presentation, a ProQOL posttest was made available to the employees via SurveyMonkey. After the data was analyzed utilizing Intellectus Statistics online computer software, results from the resiliency training on ProQOL scores was disseminated to clinical staff, stakeholders, and the union representative.

Model Framework
The John Hopkins Nursing Evidence-Based Practice (JHNEBP) Model was utilized to frame this EBP project, and its tools were referenced throughout the course of the project. Developed by a joint team of clinical nurses and nurse researchers at Johns Hopkins Hospital and School of Nursing, the JHNEBP Model "is a powerful problem-solving approach to clinical decision-making and is accompanied by user-friendly tools to guide individual or group use" (Newhouse et al., 2007). The model utilizes an organized, well-formulated, three-step process to guide practice improvements referred to as the PET (Practice Question, Evidence, and Translation) process (Melnyk & Fineout-Overholt, 2019). The JHNEBP model works well with this mental health-focused project because it acknowledges both internal and external influences on an individual and their practice; additionally, the model incorporates a humanistic approach by valuing others' opinions, perspectives, and experiences which are invaluable in working with the psychiatric provider population.

Methods
The ProQOL is the most frequently utilized scale measuring the positive and negative outcomes of working with individuals who experienced stressful events and is comprised of a 30-item Likert scale questionnaire (see Appendix H) with subscales for compassion satisfaction, burnout, and CF with good reliability and validity (Stamm, 2010). Within the ProQOL, the sum of individual scores of 22 or less indicate low levels, scores between 23 and 41 indicate moderate levels, and scores greater than 42 indicate high levels of whichever subscale is being measured. Pretest and posttest scores were gathered, entered into Intellectus Statistics online computer software, and mean scores were compared between subscales utilizing a t-test.

Results
The compassion satisfaction subscale yielded mean scores of 39.23 preintervention (n=13) and 39.90 post-intervention (n=10) (see Figure 1). Pre-intervention mean compassion satisfaction scores fall into moderate levels with 5 individuals with moderate levels and 8 individuals with high levels, and post-intervention scores also fall into moderate levels with 6 individuals with moderate levels and 4 individuals with high levels. In order to determine whether compassion satisfaction scores may have been created via a normal distribution, a Shapiro-Wilk test was conducted for pre-and postintervention scores (Razali & Wah, 2011) Table 1. The burnout subscale yielded mean scores of 21.92 pre-intervention (n=13) and 21.90 post-intervention (n=10) (see Figure 1). Pre-intervention mean burnout scores fall into the upper end of low levels with 8 individuals with low levels and 5 individuals with moderate levels, and post-intervention scores also fall into low levels with 5 individuals with low levels and 5 individuals with moderate levels. The burnout subscale yielded mean scores of 21.92 pre-intervention (n=13) and 21.90 post-intervention (n=10). In order to determine whether burnout scores may have been created via a normal distribution, a Shapiro-Wilk test was conducted for pre-and post-intervention scores (Razali & Wah, 2011) The results are displayed in Table 2. The STS subscale yielded mean scores of 19.15 pre-intervention (n=13) and 18.10 post-intervention (n=10) (see Figure 1). Pre-intervention STS scores fall into low levels with 10 individuals with low levels and 3 individuals with moderate levels, and postintervention scores also fall into low levels with all 10 individuals with low levels. The burnout subscale yielded mean scores of 21.92 pre-intervention (n=13) and 21.90 postintervention (n=10).In order to determine whether burnout scores may have been created via a normal distribution, a Shapiro-Wilk test was conducted for pre-and postintervention scores (Razali & Wah, 2011) Table 3.

Figure 1
Pre and Post-Intervention ProQOL Scores

Cost Benefit Analysis
As previously stated, the annual economic cost associated with burnout related to provider-patient alliance, reduced consumer satisfaction, and poorer patient outcomes that directly result from those issues.

Discussion
The long-term physical and mental health sequalae caused by prolonged STS, CF, and burnout are significant throughout the medical community, however, providers within the VHA acute psychiatry units are faced with a greater risk than most other settings (Garcia et al., 2015). This EBP project sought to improve resilience against STS, CF, and burnout amongst VHA acute psychiatry providers through an educational class emphasizing the risk these providers face, explaining the impact of these conditions, and introducing evidence-based solutions including LKM and the PRMA. A ProQOL scale was administered before and 30 days following the training. Although there was a slight decrease noted in STS post-intervention scores (Figure 1), unfortunately these outcomes where not clinically significant and can conceivably be attributed by loss to follow-up, however, the project did demonstrate a noteworthy presence of moderate levels of compassion satisfaction and burnout consistent with characteristic prevalence noted during the literature review. This project underscores the importance of holistic health and the benefits of self-care and is congruent with past studies showing STS, CF, and burnout are challenging problems to address. Though psychiatric healthcare workers theoretically possess the individual or organizational tools necessary to increase resilience, whether or not these providers effectively utilize these strategies to mitigate STS, CF, and burnout is largely unknown.
Limitations to this EBP project include small sample size, short duration of measure, and the inclusion of only select occupational roles within the group. A need for this training was agreed upon by the co-section chiefs of the VHA acute psychiatry units whose feedback indicated a subjective feeling of success after the training, however, it may be possible the intervention experienced poor attendance and a perception of burden upon the psychiatric providers' busy schedules due to the factors influencing STS, CF, or burnout, and these effects may have played a part in subverting the intervention as shown to be the case in past studies (Gilbody et al., 2006). The intervention recruited physicians, residents, and nurse practitioners for the educational training, however, clearly other roles such as social workers, therapists, and nurses experience the effects related to burnout, may have benefited from the training, and could have increased the sample size for a more accurate analysis of the intervention's effects.

Implications
The utilization of MBSR, LKM, and the PRMA have showed promise in improving ProQOL scores in past studies, and the results of this EBP project correspond with a systematic review and meta-analysis stating individual-tailored interventions for burnout produce only small benefits; however, these ProQOL outcomes may benefit further through the use of adjunct institutional-directed approaches (Panagioti et al., 2017) which demonstrates the need for the VHA to take measures at addressing the pervasive issues of CF, STS, and burnout within their ranks. Although this intervention failed to produce significant changes to ProQoL scores amongst the VA's acute psychiatric provider population, burnout remains a pervasive issue, and further individual and/or institutional interventions are warranted.