BACKGROUND
Care of patients after a trauma requires the engagement of the emotional, physical, and mental attributes of the health care provider. With understaffing, overcrowding, decreased available resources (Hooper et al., 2010; Wentzel & Brysiewicz, 2017), and the coronavirus pandemic, many health care professionals are experiencing a loss of empathy, exhaustion, and frustration; this is known as compassion fatigue (Joinson, 1992). Compassion fatigue impacts physicians, nurses, nursing assistants, respiratory therapists, emergency medical technicians (EMT), and EMT techs [employed by the study site's emergency department (ED)]-all of whom care for victims of trauma or may experience a patient death. Health care providers impacted by compassion fatigue have reported feeling ineffective in the workplace-resulting in low-quality care, diminished staff morale, compromises in patient safety, and resignation (Al-Barmawi et al., 2019; Cocker & Joss, 2016; Potter et al., 2013; Wentzel & Brysiewicz, 2017).
Prior to the pandemic, compassion fatigue was reported in over 25% of trauma nurses (Hinderer et al., 2014) and over 50% of ED and intensive care unit (ICU) nurses (Hooper et al., 2010,Hunsaker et al., 2015). Health care professionals may continue to experience psychological distress, burnout, and compassion fatigue due to the high-pressure and high-risk scenarios resulting from large numbers of patients experiencing dire outcomes (Alharbi et al., 2020), especially related to the coronavirus pandemic. Unfortunately, health care leadership inconsistently addresses compassion fatigue, as staff may be encouraged to be resilient and normalize work-related stress (Flarity et al., 2013).
Workplace interventions suggested to reduce compassion fatigue have included mindfulness, exercise, counseling, bereavement sessions, and structured debriefing (Boyle, 2011). Structured debriefing sessions are a nonthreatening and low-cost solution that allows participants to learn from unexpected outcomes, identify opportunities for improvement, and promote well-being (Campbell et al., 2016; Rivera-Chiauzzi et al., 2016; Schmidt & Haglund, 2017). Debriefings are a dialogue among multidisciplinary teams following a critical incident or high-stress situation (Shore, 2014). Traditionally, they are structured and last from 10 to 60 min, where health care leaders guide the team to discuss why events happened and explore the implications regarding communication, responsibilities, situation awareness, workload, and corrections in errors (Patient Safety Network, 2019; Sundheim, 2015). Debriefings have been an effective strategy in reducing compassion fatigue by allowing individuals to share their grief and stress with other colleagues (Berg et al., 2016; Browning & Cruz, 2018; Hammerle et al., 2017), resulting in greater coping skills, increased resilience, and improved social support (Schmidt & Haglund, 2017; Sullivan et al., 2019). Debriefing sessions have also identified gaps in staff communication, systematic issues, and educational needs (Gardner, 2013; Shore, 2014) and are supported by The Joint Commission (Campbell et al., 2016) as a tool for adverse-event investigation. However, it is unclear whether debriefings may help minimize compassion fatigue, particularly in emergency and critical care settings when patient death has been experienced.
KEY POINTS
* Trauma health care professionals working in high-stress environments are at risk for compassion fatigue.
* Structured debriefing sessions allow health care providers to express emotions and develop greater coping skills.
* Implementation of structured debriefing sessions is a feasible strategy to address compassion fatigue.
OBJECTIVE
To investigate the feasibility of a 12-week pilot of structured debriefing sessions and its impact on compassion fatigue experienced by emergency and intensive care health care professionals after patient death.
METHODS
Research Study Design
The Ottawa Model of Research Use by Logan and Graham (1998) was the conceptual model that guided the planning and implementation of this study. The Ottawa Model's six elements of research dissemination into practice and methodology assisted in identifying stakeholders, overcoming barriers, tracking progress, and evaluating the impact of the debriefings throughout the study implementation (Logan & Graham, 1998) (see Figure 1).
This study was conducted using a preintervention/postintervention design, where debriefing sessions were implemented once patient death was reported, to determine whether debriefing participation impacted health care professionals' compassion fatigue. The study took place in the ED and ICU of an urban, academic, 300-bed, Midwest Level II trauma center from March 1, 2021, to May, 31, 2021. The preintervention survey and participant education period began 2 weeks prior to the start of the intervention, with structured debriefings implemented for 12 weeks, and then a 2-week postintervention survey administered prior to the close of the project (see Supplemental Digital Contents 1 and 2, available at: http://links.lww.com/JTN/A52 and http://links.lww.com/JTN/A53). A validated instrument to measure compassion fatigue, the Professional Quality of Life Measure (ProQOL, version 5), was used for both surveys (Stamm, 2010). In addition, seven to eight free-text questions were asked related to age, professional role, professional experience, employment status, and the number of debriefing sessions attended. Participants were then invited to complete a preintervention survey after the project coordinator's in-person visit and educational presentation. Incentives were provided; for preintervention survey completion (pens, candy, and mini-hand sanitizers), each department and debriefing implementors received cookies in the postintervention.
Participant Inclusion and Exclusion
All eligible participants (n = 96) (nurses, nursing assistants, EMTs, EMT techs, and respiratory therapists) were permanently employed in the ED and ICU, 18 years of age, and English speaking. We identified 14 debriefing implementors, which included department managers and assistant managers (n = 5), nursing supervisors (n = 8), and spiritual services (n = 1). The study ICU also included the step-down unit, as the staff flexed between both units daily. All respiratory therapists were included because they are responsible for coverage in the ED and ICU. EMTs and EMT techs were only employed in the ED. Emergency and ICU physicians, residents, and medical students were excluded from the data collection and analysis; however, they (and any other clinicians) were invited to participate in the debriefing sessions.
Intervention Description
Handbook Design
The study design concept, implementor's handbook, and use of the three debriefing questions were adapted (with permission) from Zajac et al. (2017). The handbook contained the study description, project coordinator's contact information, data pages (20 pages), three questions to address at each debriefing, and space for postdebriefing comments. Each data collection page required the implementor to record the date, time, department, number of participants and their profession, and the debriefing session length. The handbooks were distributed after the compassion fatigue education and debriefing presentations were completed (approximately 2 weeks prior to the intervention start and collected at the project close) by the project coordinator (see Implementor's Handbook Pages, Supplemental Digital Content 3, available at: http://links.lww.com/JTN/A54).
Debriefing Questions
The three questions asked at every debriefing were structured around the relationship-based care model by Koloroutis (2004). The model states that there are three types of relationships pertinent between the caregiver and the patient: relationship with patients and families, relationship with colleagues, and relationship with self. Therefore, the three debriefing questions for this study were:
1. How did you help the patient or family through this transition?
2. What example of collaboration was most noteworthy during this patient experience?
3. What impact will this patient's death have on you?
Debriefing Implementation
During the 12-week intervention, project implementors offered structured debriefing sessions once patient death was reported. Each debriefing session took place in the perspective department, ideally outside of the patient care area, and all participating health care professionals were present within the same shift. If multiple patient deaths occurred each day, the implementor's goal was to conduct at least one debriefing session during those 24 hr. All participants received study information and compassion fatigue education materials both in-person, in print, and electronically (via hospital email) during the preintervention phase and were informed that their participation was voluntary during all phases of the study implementation.
Debriefing Criteria
All implementors were notified via page or phone call by the ED or ICU staff when patient death occurred. Given that the circumstances surrounding trauma and patient death vary with each patient encounter, the decision to implement a debriefing session was at the implementors' discretion.
ProQOL Survey
The ProQOL survey consisted of 30 questions regarding the participants' emotions and reactions in the last 30 days, with Likert responses on a scale from 1 (never) to 5 (very often). Survey analysis, interpretation, and calculations were provided by the Concise ProQOL Manual 2010 (Stamm, 2010). This survey was chosen due to its widespread use in compassion fatigue literature and diversity of clinical specialties, as seen in the literature reviewed for this study (Berg et al., 2016; Hooper et al., 2010; Potter et al., 2013; Sullivan et al., 2019; Zajac et al., 2017), suggesting good reproducibility. However, the psychometric evaluation of the ProQOL survey continues to evolve, indicating areas for improvement and strengths behind the use of the tool when applied to health care workers and caregivers (Hagan, 2019; Hemsworth et al., 2018).
The scale interpretation for compassion satisfaction (<23 indicated a low quality of life or satisfaction in the participants' profession) was calculated independently. In contrast, the final compassion fatigue score was a composite score, incorporating the burnout and secondary traumatic stress scores (scores below 22 were considered low, 23-41 moderate, and >=42 were high) (Stamm, 2010). Cronbach [alpha] scores of 0.88 (compassion satisfaction), 0.75 (burnout), and 0.81 (secondary traumatic stress) were previously reported by Stamm (2010). Cronbach [alpha] values for this sample were 0.78 (preintervention and postintervention). The preintervention and postintervention surveys also included demographic questions regarding a participant's age, professional role, years of employment at the project site, professional experience, education, and employment status during the pandemic.
Statistical Analysis
The project coordinator conducted data analysis using the StataIC 16 statistical software program (Stata Corp, College Station, TX). Univariate descriptive statistics and independent t tests (p < .05) were used to determine the effectiveness of the debriefing intervention, baseline, and analysis of differences between departments and professions, and compassion fatigue subcategories. Further analysis using [chi]2 was completed to determine the significance between education, years of professional experience, employment during the pandemic, professional role, duration of employment at the project site, and department participation. The Institutional Review Boards at the University of Michigan and McLaren Health Care reviewed this research study and deemed it was exempt from further human subject oversight prior to the study implementation.
RESULTS
Preintervention
Of the eligible participants (n = 96), 31 (32%) health care professionals completed surveys in the preintervention phase (M = 39.4, SD = 13.3); this included 14 (45%) ED employees, 10 (32%) ICU employees, and 7 (23%) respiratory therapists. Nineteen (68%) were registered nurses and 2 (7%) nursing assistants, with no EMTs or EMT techs reported. Of the total preintervention participants, 22 (77%) were employed prior to or during the pandemic (March 2020), 16 (59%) had attained a 4-year degree or higher, 17 (61%) had 5 years or more of professional experience, and 17 (68%) had less than 5 years of employment at the project location (see Table 1).
Postintervention
In the postintervention sample, 25 (26%) health care professionals completed surveys (M = 37, SD = 10.7); this included seven (28%) ED employees, 15 (60%) ICU employees, and three (12%) respiratory therapists. The postintervention sample comprised 18 (75%) registered nurses and three (12%) nursing assistants, including one EMT tech. Nineteen (86%) had worked at the project location prior to and during the pandemic, 13 (72%) had attained a 4-year degree or higher, 14 (61%) had 5 years or more of professional experience, and 13 (59%) had less than 5 years of employment at the project location (see Table 1).
Debriefing Sessions
Fifty-two patient deaths were reported hospital-wide during this study's 12-week intervention. Of the 52 patient deaths, 20 debriefing sessions were conducted and attended by 56 health care professionals; these 20 debriefing sessions represented 38% of all deaths hospital-wide during the study period. The implementors reported spending an average of 14 min (5-45 min) for each debriefing session. Registered nurses were the majority of participants (approximately 1-5) per session, along with EMT techs or nursing assistants (n = 8) and physicians (n = 4). The implementors reported no EMT technicians or respiratory therapists during any of the debriefing sessions; however, three respiratory therapists indicated that they attended at least one session on their postintervention survey.
Of the completed postintervention surveys, 10 (45%) of participants had attended one or more debriefing sessions, with half (n = 5) of the participants attending three to seven sessions. Twenty-two (55%) participants indicated that they did not attend any debriefings during the project implementation. In addition, some participants added free-text comments regarding the impact of debriefing on themselves or coworkers with varying responses from "No" to "Yes, it helped a little." Postdebriefing implementor comments-communicated to the project coordinator and written on the implementor handbooks-suggested that participants were open to expressing their grief, frustrated by the large numbers of death due to the pandemic, feeling emotional fatigue and in need of support, and were exposed to more ethical and moral dilemmas in the workplace.
Prior to the debriefing intervention, the average compassion satisfaction scores (M = 37.7, SD = 6.1) indicated that the staff had a moderate level of job satisfaction and felt valued by their colleagues in the workplace (Stamm, 2010). Low burnout (M = 24.3, SD = 6.7) and secondary traumatic stress scores (M = 23.1, SD = 6.7) indicated that the staff felt effective in their workplace and that job-related stress and trauma had a lesser impact on their personal and professional lives (Stamm, 2010). No significant differences were seen when compared with postintervention burnout (M = 25.5, SD = 5.4, p = .47), secondary traumatic stress (M = 23.9, SD = 5.6, p = .99), or compassion satisfaction (M = 36.8, SD = 6.4, p = .61) (see Table 2) among the preintervention and postintervention samples. Independent t tests were conducted comparing all departments and professions and the compassion fatigue subcategories (i.e., burnout, secondary traumatic stress, and compassion satisfaction) (see Table 3), with no significant differences noted between the preintervention and postintervention groups. [chi]2 analyses also demonstrated no significance when comparing all professions (p = .47), employment during the pandemic (p = .36), and professional experience (p = .36) between the preintervention and postintervention groups.
DISCUSSION
This study examined the feasibility of implementing structured debriefing sessions with health care professionals and its impact on compassion fatigue when patient death was experienced in the ED and ICU trauma settings. Although there were no significant differences in compassion fatigue preintervention or postintervention, the participants did indicate high levels of compassion satisfaction with moderate levels of burnout and secondary traumatic stress.
Previous trauma and debriefing research largely focused on physician, nursing, or specialty-specific debriefing sessions (Berg et al., 2014; Tyler et al., 2021). Although this research has helped identify many contributing factors reducing compassion fatigue, research addressing the multidisciplinary trauma team members, compassion fatigue, and patient death has not been well studied. Since trauma care can impact the entire multidisciplinary team and compassion fatigue can impact all health care professionals, it is vital to understand how debriefing can help support all health care professionals. Thus, a key strength of our study was the inclusion of multiple types of health care professionals-nurses, nursing assistants, respiratory therapists, EMTs, and EMT techs-and our ability to demonstrate the feasibility of a debriefing process with a multidisciplinary team.
Our findings support the feasibility of conducting structured debriefing interventions in an urban acute care hospital. In 12 weeks, 20 debriefings were implemented that covered nearly half of the patient deaths at the study site. The project team also received surveys from approximately 30 participants preintervention and postintervention-a moderate sample size given the single-center nature of the study. We acknowledge that this sample size is small, and we are underpowered to identify significant differences between groups. Thus, we encourage future researchers to consider presurvey and postsurvey pairing or other incentives to ensure a larger sample and fully powered analyses.
Despite our study's limitations, we received qualitative feedback from attendees and implementors about the feasibility and perceived effectiveness of the debriefings. Participants and implementors indicated the debriefings were helpful. Participants who attended multiple debriefings (between 3 and 7) reported that it helped process their work-related stress, suggesting a possible dose response; greater value may be attained from participating in multiple sessions or over a longer period. In addition, the timing of our study's debriefing intervention (1-year postpandemic) may have caused low debriefing attendance and survey completion. Furthermore, by March 2020, health care professionals were saturated in work-related and personal stress due to the high patient mortality and lack of resources. The pandemic has become a chronic stressor that may have impacted our ability to identify any benefit from a single debriefing session. Due to limitations on participant time, we also did not collect detailed data on all the patient deaths and associated debriefings. Considering all these factors, redesigning the methods of the debriefing interventions in frequency, place, time, choice of leadership implementors, and more detailed information about the patient death or precipitating event may have produced greater participation, larger sample size and enhanced our ability to describe the merits of debriefing; we recommend revising to address this in future work.
We acknowledge that the emotional impact of trauma and patient death results in various stress responses from health care professionals. In our study, the decision to offer a debriefing was at the discretion of the study implementors. However, debriefings may be more consistently implemented or used if all health care professionals can request it. Processes like rapid response-where any individual health care team member can call a rapid response-should be available in all hospitals to support their health care providers' well-being. Thus, we do not believe criteria for when to offer a debriefing session are warranted. Nonetheless, health care leadership should continue to offer debriefings as a sign of recognition and consistency that their staff's well-being is a priority.
Notwithstanding a possible dose-effect, debriefings may have other intended benefits, such as improving teamwork or feeling the organization's support, that we were unable to measure in our study. This study demonstrated that implementing debriefings is feasible, and our results can be used to inform future debriefing interventions.
LIMITATIONS
This study was conducted 1 month after a new hospital-wide electronic medical record implementation; this may have contributed to the small sample size and willingness to participate in debriefing sessions. Also, department nurse managers and assistant managers were included as debriefing implementors, which may have influenced debriefing attendance. There were no EMTs who participated in our surveys. Only one postintervention survey was completed by an EMT tech, therefore limiting the applicability of our findings for these professions. Three respiratory therapists did indicate their debriefing attendance; however, they were not identified by implementors in their handbooks. This error may have occurred when the implementors did not verify the profession of all people in attendance at the debriefing session. Due to high employee turnover and low staff retention, it was infeasible to link the preintervention and postintervention surveys; this may have also contributed to missing data in our survey analysis. Future studies should consider pairing the participants' surveys, as this creates greater clarity in intervention effectiveness and allows for more precision in analysis.
CONCLUSIONS
Balancing self-care and caring for others is the largest hurdle for health care professionals, especially those in high-intensity settings like trauma, emergency, or critical care. This pilot study supports the feasibility of structured debriefing sessions in an urban, acute care setting for all health care professionals. We must continue to develop opportunities to emotionally support our multidisciplinary trauma team so that they may continue to provide high-quality care for patients, families, and themselves.
Acknowledgments
The author would like to acknowledge the staff at McLaren Oakland Hospital and the University of Michigan School of Nursing for their support and advisement throughout this project.
REFERENCES