A Statistical Learning Approach to Evaluate Factors Associated With Post-Traumatic Stress Symptoms in Physicians: Insights From the COVID-19 Pandemic

Abstract

Physicians facing the COVID-19 pandemic are likely to experience acute and chronic, and often unpredictable, occupational stressors that can incur post-traumatic stress symptoms (PTSS), prevention of which is of utmost importance to enhance healthcare workforce efficiency. Unlike previous studies, in this paper we developed a generalized data-driven framework to generate insights into the complex, nonlinear associations of cognitive/occupational factors with physicians' PTSS-risk. Data were collected from practicing physicians in the 18 states with the largest COVID-19 cases by deploying a cross-sectional, anonymous, web-based survey, following the second COVID-19 peak in the US. Analyses revealed that physicians directly treating COVID-19 patients (frontline) were at higher occupational risk of PTSS than those who didn't (secondline). We implemented a suite of eight statistical learning algorithms to evaluate the associations between cognitive/occupational factors and PTSS in frontline physicians. We found that random forest outperformed all other models, in particular the traditionally-used logistic regression by 6.4% (F1-score) and 9.6% (accuracy) in goodness-of-fit performance, and 4.8% (F1-score) and 4.6% (accuracy) in predictive performance, indicating existence of complex interactions and nonlinearity in associations between the cognitive/occupational factors and PTSS-risk. Our results show that depression, burnout, negative coping, fears of contracting/transmitting COVID-19, perceived stigma, and insufficient resources to treat COVID-19 patients are positively associated with PTSS-risk, while higher resilience and support from employer/friends/family/significant others are negatively associated with PTSS-risk. Insights obtained from this study will help to bring new attention to frontline physicians, allowing for more informed prioritization of their care during future pandemics/epidemics. INDEX TERMSPost-traumatic stress symptoms (PTSS); depression and burnout; COVID-related damaging factors; resilience and social support; nonlinear relationships; predictive analytics.

Using a Classifier Ensemble for Preventing Burnout in University Students: A Study Case in Valparaíso

“…A study of the Post-Traumatic Stress Symptoms (PTSS) in physicians on the frontlines of COVID-19 was presented in [12]. The objectives of the study were (1) to evaluate the symptoms of PTSS among frontline physicians compared to second-line ones, (2) to identify the higher-risk group among them, (3) to predict the PTSS risk in the higher-risk group using ML algorithms, and (4) determine possible patterns between the predictors.…”

“How Was Work Today?”*

“…Research has demonstrated that depersonalization is the most serious aspect of burnout in terms of its impact on functioning (4) and that exhaustion, on its own, does not constitute burnout, as evidenced by its quick reversal when staff are provided with more rest (7). Finally, although the PTSD scale used cannot be regarded as diagnostic, the findings did suggest that these symptoms were elevated, as they have been found to be in other large samples of critical care workers in relation to working during the pandemic (8).…”

“…Given the predominance of cardiovascular dysfunction and coronary anomalies identified in early reports of MIS-C (12), much of the initial literature describing postdischarge outcomes following MIS-C appropriately focused on cardiac outcomes. Reassuringly, studies evaluating patients 6 to 12 months following MIS-C have demonstrated normalization of cardiac function, resolution of coronary dilation or aneurysms, and no signs of persistent myocardial inflammation in most patients (2,4,8,9). Activity impairment and exercise intolerance have been more commonly reported in the months following hospitalization for MIS-C (4,6).…”

“…A number of new insights related to this dilemma are provided in another recent study, which used statistical learning algorithms to examine risk and protective factors associated with symptoms of PTSD in n = 1,017 physicians' , after the second wave of COVID-19 in the United States (8). The authors found that around 30% of their sample exhibited probable or borderline PTSD and that certain PTSD symptoms discriminated between those who were frontline in the pandemic and those who did not treat COVID patients directly.…”

“…Among the numerous commendable elements of the robust multicenter research collaborations that grew out of the COVID-19 pandemic (2-7) was the foresight to prioritize the study of long-term outcomes from the beginning. Within the first 18 months of the pandemic, literature was already beginning to emerge describing postdischarge outcomes for children surviving acute COVID-19 and multisystem inflammatory syndrome in children (MIS-C) (8)(9)(10)(11).…”