Resiliency and Mental Health in the ICU

The coronavirus disease 2019 (COVID-19) pandemic has left a profoundly changed critical care workforce in its wake. The practice of critical care is demanding in normal times. Before the pandemic, guidelines for hospital disaster responses and public health emergencies assumed a time-limited nature. Concepts such as conventional, contingency, and crisis standards of care are organizational tools that allow health systems to adapt to an emergency and provide the best possible care to the greatest possible number of patients (1). The prolonged nature of the COVID-19 emergency prevented many of these notions from working as intended; instead of a time-limited disaster, the recurrent waves and surges led to a “chronic contingency,” where shortages of staff, supplies, and space became routine rather than exceptional (2).

Although our professions may select individuals with higher degrees of resiliency than the general population, we conversely face high rates of burnout, depression, substance abuse, and risk for suicide (3, 4). These challenges predate the pandemic and have only grown in magnitude since 2020 (5, 6), with rates of depression and post-traumatic stress disorder (PTSD) as high as 50% among healthcare professionals (HCPs) (7). The reasons for this degradation of HCP mental health are not difficult to understand. We all recall the banners and posters referring to HCPs as heroes, but ICU staff worked with inadequate personal protective equipment, haphazard training, feckless leadership, and a political and media environment that favored sensationalism over science (the “infodemic”). Between January 2020 and May 2021, more than 115,000 HCPs died of COVID-19 around the world, with many of these deaths likely preventable with better planning (8). If this is how society treats its heroes, one can only imagine how it treats its villains.

In this issue of the Journal, Azoulay and colleagues (pp. 573–583) report on the results of a cross-sectional survey of HCPs working in French ICUs (9). Using validated tools, the authors measured respondents’ responses to the Connor-Davidson Resilience Scale (CD-RISC-10) during the Omicron wave of COVID-19 in late 2021. They then correlated the results of the CD-RISC-10 with self-reported symptoms of anxiety, depression, and PTSD, as well as demographic and professional factors, to assess whether these features of resiliency were associated with improved mental health (9).

A concern with a study of this nature is one of tautology: resilient people will have fewer mental health symptoms because that is the definition of resiliency. The CD-RISC-10 uses a validated questionnaire with 10 items regarding one’s perceived ability to cope with stressful situations, such as adapting to change, staying focused under pressure, and handling unpleasant feelings. These items are rated on a Likert scale with scores ranging from 0 (meaning “not true at all”) to 4 (“true nearly all the time”) (10). Participants’ responses on the Hospital Anxiety and Depression Scale and the Impact of Events Scale-Revised were then used to measure symptoms of anxiety and depression as well as PTSD, respectively. As measured by the CD-RISC-10, features of resiliency appear to be protective against mental health symptoms, as expected, but disentangling the definitions and causality can still be difficult, despite this correlation.

Other features of this study merit further comment. This study evaluated an experienced cohort of ICU professionals, with 93% working in prior COVID-19 waves. Although this may suggest a high degree of resilience, it also raises the possibility of a survivor bias. This group may be highly resilient because they were the ones left standing; less resilient staff may have left the ICU or, indeed, the profession entirely before the Omicron wave (11).

The concept of post-traumatic growth is an interesting one. In earlier waves, resilience may improve after enduring a traumatic event, although the prolonged and repeated nature of pandemic surges may have worn away the benefits from earlier waves. This dichotomy can be seen in the present work: experience with the care of patients with COVID-19 was associated with greater resilience, but caring for more than 10 patients with COVID-19 who died was associated with decreased resilience. There is an inflection point to be found here between experience and tragedy, perhaps influenced by working with patients’ families in end-of-life decision making (as a surrogate for maintaining humanistic relationships). Social support from one’s family, friends, colleagues, and the broader society may help further reinforce this post-traumatic growth and support HCP resilience (12).

The difference in reported resilience between men and women in this cohort also needs further exploration. The authors note that men and women may respond to these survey instruments differently, to suggest artificially greater resilience in one sex over another. Another consideration, however, is the differential distribution of labor and responsibilities outside of the hospital, which may further burden female HCPs. Globally, the great majority of HCPs are women, including 75% of HCPs in the United States (and 88% of nurses) (13). Paid time off, mentorship, childcare, and flexible scheduling are not issues exclusively aimed at supporting women in health care, but any interventions to improve overall HCP resilience must address them.

Calls for increased resiliency by HCPs can be a way for leaders to shift the burden away from institutional failures onto the HCPs themselves; instead of offering improved pay, flexible hours, and safe workplaces, they can instead offer yoga classes and online training (14). Significantly, only 8% of this highly experienced cohort felt adequately supported by their institutions, a woefully low number after enduring nearly 2 years of the pandemic at the time of the survey. We are, however, professionals, and we cannot neglect our responsibilities for maintaining our personal readiness: adequate sleep, nutrition, exercise, professional development, and the like. Although it is our responsibility to do these things, it is the responsibility of health system leadership to make them possible. Azoulay and colleagues have made an important contribution to our understanding of HCP resilience and mental health. Let us hope that we can learn their lessons before the next emergency.