COVID-19 and Lifetime Experiences of Trauma, Moral Injury, and Institutional Betrayal among Healthcare Providers

Abstract

Objective:

To increase understanding regarding healthcare provider experiences with psychological trauma, moral injury and institutional betrayal, both over the lifetime and during the COVID-19 pandemic.

Methods:

The study employed a cross-sectional design to understand traumatic experiences, moral injury, and institutional betrayal among medical and mental health providers. Participants were asked to identify an index trauma, and experiences were coded qualitatively using categories for traumatic events, moral injury, and institutional betrayal.

Results:

Results revealed that experiences of trauma, moral injury, and institutional betrayal were common in relation to the pandemic, as were pre-pandemic histories of traumatic exposures. Findings indicate that trauma exposure was a work hazard for healthcare providers during the pandemic, which could result in negative long-term mental health outcomes.

Conclusions:

Future research is needed to explore potential long-term negative outcomes among healthcare providers.

Introduction

Civilian healthcare providers faced unprecedented workplace challenges during the Coronavirus Disease (COVID-19) pandemic. While exposure to illness and death are routine healthcare experiences, extraordinary numbers of deaths occurred during the pandemic.¹ In comparison to non-pandemic times, healthcare workers have likely been exposed to higher rates of psychological trauma due to high death rates, severity of illness, and concerns about potential infection of self and/or others. Early in the pandemic, providers were further challenged by limited resources resulting in difficult choices regarding care rationing.²

Trauma exposure is the driving factor in the development of posttraumatic stress disorder (PTSD).^3,4 The Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, defines trauma as “exposure to actual or threatened death, serious injury, or sexual violence,”⁴ in which “exposure” includes directly experiencing the traumatic event, witnessing the traumatic event, learning about the traumatic event happening to a close family member or friend, or experiencing repeated or extreme exposure to aversive details of the traumatic event.⁴ Traumatic stressors experienced by healthcare workers include, but are not limited to: treating trauma-related injuries; witnessing or experiencing physical assault by patients; caring for terminally ill patients; and seeing patients die.^5,6 Criteria for PTSD also includes four clusters of symptoms such as intrusion (e.g., frequent, unwanted memories of the trauma), avoidance (e.g. avoiding thoughts, feelings, people, or places related to the trauma), alterations of cognition or mood, and increases in arousal or reactivity. To meet diagnostic criteria, these symptoms must cause clinically significant distress and persist for at least one month.⁴

Even prior to the onset of the COVID-19 pandemic, healthcare providers were at elevated risk for PTSD^6–9. Studies examining PTSD symptoms in healthcare workers via self-report and across different settings have found probable rates of PTSD from 8% to 24%.^6–9 Trauma exposure is also associated with other mental health disorder symptoms,¹⁰ and the presence of mental health symptoms are associated with increased suicide risk.^11,12 Extant literature indicates that prior to the pandemic healthcare providers were at increased risk for suicide.^13–15

In addition to trauma exposures during the COVID-19 pandemic, healthcare workers may have also experienced potentially morally injurious events (PMIEs). There is no consensus in the literature regarding the definition of moral injury.^16–22 The functional model of moral injury posits that moral injury responses are expected reactions to these events, and should not be conceptualized as a responses indicative of a mental health disorder.¹⁹ This model defines moral injury as a response that occurs due to the inability to manage moral pain, resulting in distress that may be spiritual, psychological or social in nature.¹⁹ This paper adopts this definition of moral injury.¹⁹

In contrast to the definition of moral injury, there is consensus regarding the definition of a PMIE, which pertains to a perceived transgressive action, either witnessed or perpetrated, or betrayal, that is counter to one’s moral code.^16–18,21 PMIEs may or may not result in moral injury. The primary literature base for moral injury is drawn from samples of Active Duty Service Members or Veterans^16,23,24; however, moral injury is a concept that has been applied to situations beyond these populations and settings. In a sample of Romanian physicians, approximately 50% experienced a PMIE during the pandemic.²⁵ Other early research during the COVID-19 pandemic suggested that mismanagement of pandemic-related workplace issues may have resulted in a higher likelihood of individuals experiencing a PMIE.²⁶

Institutional challenges may create environments that lead to both PMIEs and/or institutional betrayal. The latter occurs when an organization is perceived to violate trust, or harm one or more members of the institution, whether through purposeful action or inaction.^27–29 Institutional betrayal has primarily been studied within the context of sexual assault and military sexual trauma.^28,30 Yet, there is a growing awareness of the role of institutional betrayal in other settings, such as healthcare.³¹ Institutional betrayal has been found to exacerbate stress responses to trauma in individuals who experienced sexual assault.^28,30 In Veterans who experienced military sexual harassment and/or assault, institutional betrayal was associated with symptoms of PTSD and depression, and prior suicide attempts.³⁰ In a sample of healthcare workers in Israel and Switzerland during the pandemic, fear of COVID-19 was associated with higher levels of perceived institutional betrayal, and institutional betrayal was associated with higher negative affect.³²

Greater understanding of the trauma, moral injury, and institutional betrayal experienced by providers may inform pandemic recovery, including efforts to prevent negative outcomes. This study sought to conduct a mixed method analysis of cross-sectional survey data regarding pre- and post COVID-19 psychological trauma, as well as pandemic-related moral injury and institutional betrayal to increase understanding regarding civilian provider experiences.

Methods

Participants were U.S. healthcare providers (N = 234) working during the COVID-19 pandemic. Recruitment strategies included distribution of the consent and survey link through social media (e.g., Twitter, Facebook, Reddit) and relevant email distribution groups. Participants were included if they identified themselves as a healthcare provider between ages 18 and 89 and provided a U.S. zip code. Colorado Multiple Institutional Review Board approval was obtained prior to the start of the study (COMIRB # 20–0773).

Data collection began in May 2020, and data were pulled for the current analysis in January 2021. Thus, data were collected prior to widespread availability of vaccines. An anonymous survey link was distributed via Research Electronic Data Capture (REDCap) tool.³³ A postcard consent form and contact information for the study team was presented on the first page of the survey. After completing the postcard consent, participants completed the baseline assessment consisting of self-report measures via REDCap. Sample demographics, including age gender, race, education, profession, work setting, average percent efforts in settings, number of COVID-19 patients that the participant cared for since the start of the pandemic, and number of COVID-19 patients that care for that died, were obtained through a demographic form created by the study team. Participants were asked about trauma via a question about DSM-V PTSD Criterion A. Specifically, participants were asked to self-report “the worst event” leading to problems following “a very stressful experience involving actual or threatened death, serious injury, sexual violence, or providing healthcare during a pandemic” that was directly experienced, witnessed, or that that they learned about happening to someone close to them (i.e., DSM-V PTSD Criterion A).⁴ Participants were asked to indicate when the event occurred and were provided an open text box to briefly describe the trauma.

Stein and colleagues’³⁴ scheme for categorizing traumatic events was used to code traumas reported in the narrative response. The scheme (2012) consists of six traumatic event categories, namely: Life Threat to Self; Life Threat to Others; Aftermath of Violence; Traumatic Loss; Moral Injury by Self; and Moral Injury by Others. As mentioned in the introduction, PMIEs are distinct from moral injury. In these coding practices, events were conceptualized and coded as moral injury by self or others as the assessment of the events asked about whether the event led to additional problems. In addition, Institutional Betrayal ²⁷ emerged as an additional theme and was added as a seventh category. Definitions for each category are provided in Table 1. These categories are not mutually exclusive, as traumas could be coded under multiple categories (e.g., sexual assault is coded as both Life Threat to Self and Moral Injury by Others as per the Stein classification).³⁴ The first and second authors discussed and coded one-third of participant responses together to promote consistency and reduce potential biases in assigning codes, and then divided and independently coded the remaining traumas. Raters met on multiple occasions to reach consensus. Additionally, raters also spot checked each other’s independently coded traumas and discussed potential discrepancies until reaching consensus to promote coding consistency and reduce potential biases. To provide context to provider quotes while at the same time protecting participant confidentiality, participant quotes were attributed as either coming from a “Medical Provider” or “Mental Health Provider.”

Table 1.

Categories of traumatic events used to code each trauma

Category	Definition
Life Threat to Self	Experiencing a perceived or actual life-threatening event
Life Threat to Others	Witnessing a perceived or actual life-threatening event
Aftermath of Violence	Being exposed to disturbing images after a traumatic event, like severely ill or injured people or dead bodies
Traumatic Loss	Witnessing or learning about the death of someone like a family member, friend, patient, etc.
Moral Injury by Self	Acting in a way that violates one’s own moral or ethical code
Moral Injury by Others	Witnessing or experiencing actions by others that violate moral or ethical codes
Institutional Betrayal	When an institution violates the trust or harms a member of the institution, whether through purposeful action or inaction

Results

The majority of the sample was female (n = 187, 79.91%), and identified as Caucasian/White (n = 195, 83.33%), obtaining a Ph.D. (n = 109, 46.58%; see Table 2 for details). The most frequent profession was mental health provider (n = 143, 61.11%) and the most frequent work setting was the Department of Veterans Affairs (n = 118, 50.43%; see Table 2). Overall, 42.73% (n = 100) of participants experienced a trauma in their lifetimes that was coded per one or more of the Stein traumatic events categories. Additionally, 18.38% (n = 43) of the sample reported an event that did not meet A1 criteria or did not provide enough information for the raters to code it. Importantly, 23% (n = 23) of those reporting a trauma described an experience related to the COVID-19 pandemic (see Table 3).

Table 2.

Demographic and Work Setting Information

	N	%	Mean	Range
Gender1
Male	33	14.10%
Female	187	79.91%
Trans	0	0.00%
Other	3	1.28%
Decline to Respond	1	.43%
Race2
Caucasian/White	195	83.33%
Black/African American	3	1.28%
Native American/Alaska Native	1	.43%
Asian	10	4.27%
Pacific Islander	0	0.00%
Multiracial	7	2.99%
Other	7	2.99%
Decline to Respond	1	.43%
Education
High School Diploma or Equivalent	2	.86%
Some College, No Degree	1	.43%
Associate’s Degree	4	1.71%
Bachelor’s Degree	25	10.69%
Master’s Degree	58	24.86%
Doctoral Degree (Ph.D.)	109	46.58%
Doctoral Degree (M.D./D.O.)	35	15.0%
Profession
Physician	27	11.54%
Nurse	27	11.54%
Respiratory Therapist	0	0.00%
Physical Therapist	8	3.42%
Occupational Therapist	4	1.71%
Mental Health Provider	143	61.11%
Nurse Practitioner	6	2.56%
Physician Assistant	1	.43%
Speech Language Pathologist	0	0.00%
Phlebotomist	0	0.00%
Technician	1	.43%
EMT/Paramedic	1	.43%
Other	16	6.84%
Work Setting
Department of Veterans Affairs Facility/Hospital	118	50.43%
Academic Medical Facility/Hospital	54	23.08%
State or Community Health Facility/Hospital	25	10.69%
Private Practice	18	7.69%
Other	19	8.12%
Average percent effort in settings
Outpatient		71.61%
Inpatient		27.67%
Number of COVID-19 patients cared for since the start of the pandemic			86.25	0–200
Number of COVID-19 patients cared for that died			1.303	0–50

¹n = 10 missing

²n = 10 missing

Table 3.

Prevalence of Trauma Types Related and Unrelated to the Pandemic

Trauma Type	Pandemic Related Traumas (n = 23)		Non-Pandemic Related Traumas (n = 77)
	n	%	n	%
Life Threat to Self	7	30.43%	18	23.38%
Life Threat to Other	14	60.87%	24	31.17%
Aftermath of Violence	0	0.00%	1	1.30%
Traumatic Loss	6	26.09%	28	36.36%
Moral Injury by Self	0	0.00%	1	1.30%
Moral Injury by Others	4	17.39%	23	29.87%
Institutional Betrayal	5	21.74%	2	2.60%

COVID-19 Pandemic Related Traumas

Among COVID-19 pandemic related traumas (n = 23), 30.43% (n = 7) were related to Life Threat to Self. One provider described how the hospital, “…took [personal protective equipment (PPE)] for the main hospital … [and] require[ed] diabetic [healthcare providers] … [see patients] face to face… despite [diabetes] being a very high-risk factor for death from COVID-19” (Medical Provider; also coded as Life Threat to Others and Institutional Betrayal). Another provider described her/his experience as a patient, stating, “Emergency surgery and hospitalization with 4 days in the ICU (including time on a ventilator)” following a COVID-19 infection (Medical Provider). The majority of COVID-19 related traumas were categorized as witnessing Life Threat to Others (n = 15, 65.21%). One provider described her/his traumatic experience, as well as detailed the stressors s/he endured daily during the pandemic:

… I have had [patients] code while in my care, later die... In general working in [a hospital] at this time in a state that is not closed or taking precautions has been stressful... Working with the public has been stressful in terms of treating the sick, educating patients and encouraging people to take precautions without being yelled at regarding ‘my rights’... (Medical Provider; also coded as Traumatic Loss)

This provider goes on to capture how evolving medical practice recommendations and being at an increased risk for infecting others also impacted her/his home life:

…During the pandemic I have mostly worried, despite using PPE, that I will bring [COVID] … home with me and spread it to a loved one… it is especially scary to think of potentially infecting a loved one and caused them to become sick and potentially die. This is what has caused me the most stress and anxiety, especially [because] our hospital has had shortages of PPE, and best practices have been constantly evolving.

Similarly, another provider indicated, “Witnessing rapid decline in patients first-hand. Watching [oxygen] saturation’s [sic] plummet with basic tasks, extreme confusion/delirium, walking by room after room in the ICU with patients proned and vented” (Medical Provider). Thus, healthcare providers experienced trauma related to both Life Threat to Self and Life Threat to Others. Some experienced this debilitating illness, and others described watching their patients become extremely ill. Notably, the threat of becoming sick and/or passing COVID-19 to others was a worry that weighed heavily upon healthcare providers.

Overall, 26.09% (n = 6) of the COVID-19 related traumas were due to Traumatic Loss. One provider described,

I had [a new] patient… [who] was not stable medically upon evaluation. The patient was calling out for help and presenting with increased confusion beyond baseline, this went on all morning of the evaluation, each time I passed the room I would hear them calling for help. Multiple providers did their part to help the individual. By midday a code was called and before I knew it the patient could not be resuscitated and pronounced dead (Medical Provider).

A common subtheme in provider accounts of trauma was the desire to help patients recover but feeling helpless to do so because of the limited interventions and rapid patient decline. This led to some providers witnessing many deaths among their patients, at much higher volumes than previously experienced.

Mental health providers also reported traumas related to the pandemic. One provider described that a “patient … shot [themselves] in the head shortly after the pandemic started” (Mental Health Provider); highlighting providers’ exposure to traumas related to mental health during the pandemic.

Circumstances such as the pandemic have the potential to bring up situations of rationed care, or other challenging situations. These difficulties can result in moral injury related to behaviors conducted by self or others. Of the traumas experienced, 17.39% (n = 4) were related to Moral Injury by Others. One provider wrote:

I work in a hospital-based nursing home and all of the patients were transferred out to turn the unit into a COVID unit… The question of which lives are more valuable was brought up repeatedly. Four patients (10%) have died so far after moving, no way to know which ones would have died without the move (Medical Provider; also coded as Life Threat to Others, and Institutional Betrayal).

One provider relayed another difficult situation, stating,

Another one of our patients traveled, contracted covid, then lied about their travel history during the course of treatment until they [were] quite obviously sick. [S/he] exposed many in the [hospital]. Then, I think [her/his] own care was compromised because [her/his] care team was so angry with [her/him] (Medical Provider; also coded as Life Threat to Self, and Life Threat to Others).

Of the sub-sample, 21.74% (n = 5) identified Institutional Betrayal. One provider relayed,

I was exposed to COVID-19 while at work and without proper PPE. I was notified by Occupational Health about a week after the exposure and was denied COVID-19 testing at the time because I had no symptoms. Meanwhile, I had been working with high-risk patient populations within the hospital. (Medical Provider; also coded as Life Threat to Self and Life Threat to Others).

During the pandemic, healthcare providers reported their institution was often not able to respond to staff issues related to exposure. Of the traumas reported, none were related to the Aftermath of Violence, or Moral Injury in Self.

Lifetime Trauma Unrelated to the COVID-19 Pandemic

Of the traumas experienced during the lifetime that were unrelated to the pandemic (n = 77; 77.00%), 44.16% (n = 34) were related to Life Threat to Self. One healthcare provider indicated, ‘[My] ex-[partner] threatening to kill me and trapping me and my… [child] in our home for over an hour” (Medical Provider; also coded as Life Threat to Others). Another healthcare provider wrote, “Mother attempting to kill me” (Medical Provider; also coded as Moral Injury by Others). Also, among the traumas unrelated to the pandemic, 20.78% (n = 16) were associated with Life Threat to Others. One healthcare provider reported “…being hit, run over, and pinned under a car” by a partner (Medical Provider). Also, many traumatic events (36.36%; n = 28) unrelated to the pandemic were coded as Traumatic Loss. One healthcare provider relayed, “My [partner] [died by] suicide… I was the one who found [him/her]” (Medical Provider). Another provider stated, “…Police officers in my community were pursuing… people involved in a [crime] and were shot at. Two died that I knew very well and one was seriously injured and almost lost his life. The last one was a friend of mine” (Mental Health Provider).

Healthcare providers also confronted situations in their lifetime related to moral injury outside the context of the pandemic. Specifically, one healthcare providers indicated s/he experienced Moral Injury by Self (n = 1; 1.30%), and several (n = 23; 29.87%) reported experiences of Moral Injury by Others. In terms of Moral Injury by Self, one healthcare provider relayed,

My [partner] and I decided to get a divorce after having the worst fight in our 10 year relationship. There was no physical violence but there was screaming, yelling, and objects were thrown and broken… I went to leave the house and fell down the stairs and hit my head. It was terrifying (Medical Provider; also coded as Life Threat to Self, and Moral Injury by Others).

In terms of Moral Injury by Others, one healthcare provider wrote, “I was choked by an ex-partner and was afraid I would die” (Medical Provider; also coded as Life Threat to Self).

Only one healthcare provider (1.30%) had an event coded as Aftermath of Violence. The healthcare provider described a harrowing situation, stating,

I witnessed racism by cops ... I watched a black man, with his hands raised above his head, get shot in the back with a pepper spray ball, then tackled by the police against my car. The cops used their baton to hit the man while arresting him… (Medical Provider; also coded as Life Threat to Others, Moral Injury by Others, and Institutional Betrayal).

Of the sample, 2.60% (n = 2) reported traumas coded as Institutional Betrayal. One healthcare provider reported “Being sexually assaulted by a [healthcare provider]” (Medical Provider; also coded as Life Threat to Self and Moral Injury by Others).

Various traumas were discussed in relation to the pandemic and lifetime trauma unrelated to the pandemic. In terms of the pandemic, traumas related to Life Threat to Self and Life Threat to Others, and Institutional Betrayal were more common. During the providers’ lifetime and outside of the COVID-19 pandemic, experiences related to Traumatic Loss and Moral Injury by Others were more common. Trauma related to the Aftermath of Violence and Moral Injury by Self were rare during and outside of the pandemic. Notably, despite the relatively brief time of the pandemic at the time of this study relative to the length of time prior to the pandemic, nearly a quarter of all traumatic experiences occurred during the pandemic, suggesting a higher likelihood of experiencing trauma among healthcare providers during this time period.

Discussion

Findings from this preliminary study suggest that healthcare workers experienced notable rates of new onset trauma, moral injury, and institutional betrayal related to the pandemic. In addition, 77% of the traumatic exposures noted occurred prior to the pandemic. Such experiences could lead to the development of mental health symptoms, and have the potential to impact long-term functional outcomes, including occupational functioning.^{10–12,35,36} Indeed, previous epidemics have resulted in poor mental health outcomes for healthcare providers. Individuals who responded to the 2003 Severe Acute Respiratory Syndrome (SARS) outbreak had significantly increased levels of workplace stress, depression, and anxiety.³⁷ Providers in healthcare settings with increased risk of SARS had higher levels of depression, anxiety, and PTSD associated with workplace stress a year later compared to those in lower-risk healthcare settings.³⁷ Major contributing factors to healthcare workers’ distress during the SARS outbreak were perceived loss of control, fear of catching the virus, and spread of the virus.^38,39 During this epidemic, heavy workload and unsafe work settings also contributed to poor psychological outcomes among healthcare workers in previous epidemics.^36,38 These risk factors are highly relevant to healthcare workers’ COVID-19 pandemic experiences.

Similar early findings are being observed in the current pandemic. Studies suggest that 50.4–56.4% of healthcare providers experienced symptoms of depression,^40,41 with higher rates of moderate to severe depression among healthcare providers working directly with COVID-19 patients.⁴⁰ Providers working directly with COVID-19 patients also experienced significantly higher levels of PTSD than those engaged indirectly with patient care.⁴⁰ In addition to trauma exposure and PTSD, healthcare workers have endorsed experiencing moral injury during the current pandemic,⁴² and experiences of moral injury have been correlated with mental health symptoms and thoughts of suicide.^23,24 Healthcare workers have also reported experiences of institutional betrayal.³² Of interest, there were significant differences in the levels of institutional betrayal experienced between samples in Switzerland and Israel, with healthcare workers in Israel reporting higher levels of institutional betrayal.³² This finding suggests differences in institutions’ responses to the same global event impact experiences of institutional betrayal. It is also possible that cultural, systematic, or financial differences also resulted in these differences.

Trauma exposure, moral injury, and institutional betrayal in Veterans are associated with poor long-term outcomes, such as PTSD symptoms and other mental health symptoms, including depression, increased suicide risk, and poor functional outcomes.^{10,12,23,24,30} Indeed, research on the impact of trauma and resulting PTSD in Vietnam Veterans indicates PTSD symptoms are still present 40 years after the traumatic event in some individuals.^43,44 Findings from Steenkamp and colleagues (2017) indicated that negative reception upon returning home and high stress in the last year were associated with increased long-term PTSD symptoms.⁴⁴ Within the context of the COVID-19 pandemic, healthcare providers have faced mixed reception from the public, as noted in a quote from a provider in this study. At times, healthcare providers have been hailed as heroes, and at other times they have experienced significant hostility. Further, a large subsample of the healthcare workforce also had notable trauma exposure prior to the pandemic. Drawing from existing literature about Veteran populations, this is a critical finding, as repeated trauma exposure is associated with increased symptoms of PTSD.⁴⁵

There are several evidence-based treatments to address mental health symptoms of PTSD, and other mental health disorders. In particular, drawing from the substantial literature in military and Veteran populations, both Prolonged Exposure and Cognitive Processing Therapy have strong evidence bases.⁴⁶ However, more research is needed to understand how to effectively implement these treatments are accessible to healthcare providers.

While this study had notable strengths, it also contained several limitations. Criterion A traumas were collected via self-report. Ideally, clinicians would have assessed multiple traumas and determined a Criterion A trauma. However, this was not feasible within the broader study design. Data for this study were collected in the earlier part of the pandemic. Thus, it is likely that healthcare providers sustained additional trauma, moral injury, and institutional betrayal that was not captured in these data. Additionally, the majority of healthcare respondents were mental health providers. It is possible that rates of pandemic related trauma would be even higher in a sample primarily composed of healthcare providers who served on the front line of direct patient care. The first and second authors coded the traumatic events. While methods were put into place to reduce potential bias such as using operational definitions of the codes from prior established literature, and having multiple discussions to ensure accuracy of coding, it is possible that bias in coding occurred.

Prior research on trauma exposure, moral injury, and institutional betrayal in healthcare workers during the pandemic has tended to examine each one of these constructs separately. When all three constructs were examined concurrently within the context of this study, some pandemic experiences were intersectional and crossed the boundaries of these different types of exposures. Future research should examine the impacts of these types of events that result in concurrent exposure to trauma, moral injury, and institutional betrayal. Future research is needed to understand the long-term impacts of pre-COVID-19 and pandemic-related traumatic exposures on mental health symptoms and long-term functioning in healthcare workers responding to the pandemic.

SMART Learning Outcomes:

• After reading this paper, the audience will be able to understand the inter-related concepts of trauma, potentially morally injurious experiences, and institutional betrayal

• Through reading this paper, readers will understand healthcare providers’ experiences of trauma, potentially morally injurious experiences, and institutional betrayal during the COVID-19 pandemic and their lifetime.