Healthcare Workers and Patient Loss During the COVID-19 Pandemic: Interaction Between Grief, Viral Anxiety, and Depression
Article information
Abstract
Objective
Coronavirus disease-2019 (COVID-19) had a significant impact on the mental health of healthcare workers. Related assessments should be included in plans for future pandemics. We investigated the connections between grief, viral anxiety, depression, and preoccupation in the context of COVID-19 in healthcare workers, to determine which factors will need to be incorporated.
Methods
A total of 267 healthcare workers who had experienced the death of at least one patient during the COVID-19 pandemic were asked to respond to questionnaires assessing grief, viral anxiety, depression, loneliness, and preoccupation with COVID-19, based on their emotional state during the 2 weeks immediately after the death. Multivariate linear regression, causal mediation analysis and structural equation modeling were used to examine the psychological processes underlying grief.
Results
Linear regression showed that viral anxiety (β=0.287, 95% confidence interval [CI]: 0.235–0.338, p<0.001) and depression (β=0.157, 95% CI: 0.073–0.241, p<0.001) had independent associations with preoccupation with COVID-19. Causal mediation analysis revealed that both viral anxiety (proportion mediated: 0.51, 95% CI: 0.37–0.66, p<0.001) and depression (proportion mediated: 0.77, 95% CI: 0.50–1.08, p<0.001) were mediators between grief and preoccupation with COVID-19. In a structural equation model, viral anxiety and depression mediated most of the effects of grief on preoccupation with COVID-19, and loneliness was a mediator between grief and depression.
Conclusion
Contingency plans for the next pandemic are being formulated. Effective measures to protect the mental health of healthcare workers should be included, and such measures should consist of assessments for grief, viral anxiety, depression, and loneliness.
INTRODUCTION
During the severe acute respiratory syndrome coronavirus-2 pandemic, the continuous high demand placed on healthcare systems resulted in significant mental health problems in healthcare workers, with serious levels of depressive symptoms and burnout [1]. Prevalence reports of anxiety ranged from 7% to 57%, and 9% to 51% for depression [2]. Loneliness was also prominent in healthcare workers [1].
Persistent exposure to patient deaths in healthcare workers can increase their workload and lead to deep grief and burnout [3], as occurred during the pandemic, as healthcare workers were tasked with comforting patients while coping with high death rates. This denied them the opportunity to process their grief and increased the risk for complicated grief [4].
Complicated grief reactions are characterized by prolonged painful emotions related to bereavement and a significant loss of function (Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition [DSM-5]), and may affect nearly 60% of nurses [5]. New support systems for healthcare workers will need to consider complicated grief, and also depression and anxiety, as complicated grief is associated with both [6]. Knowledge of the relationship between grief, depression and anxiety will be invaluable when designing such systems.
Preoccupation with the pandemic has been linked to higher levels of functional impairment in healthcare workers [7]. After studying the interactions of viral anxiety and depression in healthcare workers during the pandemic [8,9], we found that both affected preoccupation with coronavirus disease-2019 (COVID-19) [10]. Therefore, grief may be associated with preoccupation with COVID-19, and viral anxiety and depression may mediate this interaction.
Loneliness is a symptom of complicated grief (DSM-5), and a significant predictor of exhaustion and worse mental health [11,12]. A study on young adults undertaken during the pandemic showed that it was associated with depression and anxiety [13]. Thus, loneliness may be a mediator between grief, depression, and anxiety.
As explained above, previous studies have reported that both viral anxiety and depression are associated with complicated grief, preoccupation with COVID-19 and loneliness. These results hint at a wider relationship encompassing all of these factors. Therefore, we aimed to identify this relationship by determining the connections between grief, viral anxiety, depression and preoccupation in the context of COVID-19. We hypothesized that (A) grief would influence preoccupation with COVID-19, and (B) the association between grief and preoccupation with COVID-19 would be mediated by viral anxiety, loneliness and depression.
METHODS
Participants
We conducted an online survey between June 2 and 11, 2022 on healthcare workers of a tertiary hospital situated in Seoul, South Korea. Medical doctors (n=1,769) and nursing professionals (n=4,680) were asked to respond if they had experienced the death of at least one patient during the past 2 years of the COVID-19 pandemic. This period represented the most extreme period of the pandemic, as monthly cases increased from under 100 to nearly 10 million (Korean Disease Control and Prevention Agency). Participation was voluntary, for which a gift coupon worth 5 US dollars was provided. The survey form was developed according to the Checklist for Reporting Results of Internet E-Surveys (CHERRIES) guidelines [14], and checked by an investigator (S.C.) prior to implementation. The CHERRIES was developed by the Journal of Medical Internet Research to improve the descriptions of web-based surveys and focuses on the sample selection process.
We collected responses from 293 (4.5%) participants, and 26 who had not experienced patients’ deaths during the past 2 years were excluded from the study. The Institutional Review Board of Asan Medical Center (2022-0740) approved the study protocol and waived the requirement of the informed consent form.
Measures
Participants were asked to recall their emotional state during the 2 weeks immediately after experiencing the death of their patient and respond accordingly.
Pandemic Grief Scale for healthcare workers
The Pandemic Grief Scale (PGS) is a self-reported rating scale measuring individuals’ grief reactions during the COVID-19 pandemic [15]. The healthcare worker’s version was revised from the original PGS by the investigator in Korean [16], and consists of 5 items which can be rated on a 4-point scale. The total score reflects the probability of a dysfunctional grief response. We translated the original English version into Korean by translation and back-translation methods. The Cronbach’s alpha for this sample was 0.842.
Fear of COVID-19 Scale
The Fear of COVID-19 Scale (FCV-19S) is a rating scale which measures an individual’s anxiety response related to the viral pandemic [17]. All seven items in the FCV-19S can be rated on a 5-point scale. A higher total score reflects a higher level of viral anxiety. We used the validated Korean version [18]. The Cronbach’ alpha was 0.889 among this sample.
Obsession with COVID-19 Scale
The Obsession with COVID-19 Scale (OCS) measures an individual’s degree of preoccupation with the coronavirus [19]. It consists of four items which can be answered on a 5-point scale. A higher total score reflects a higher level of preoccupation. We used the Korean version which has been validated [20]. The Cronbach’s alpha among this sample was 0.832.
Patient Health Questionnaire-9 items
The Patient Health Questionnaire-9 items (PHQ-9) is a self-report rating scale which measures an individual’s depressive symptoms [21]. It consists of nine items rated on a 4-point scale, and a higher total score indicates a higher level of depression. The validated Korean version was used in this study [22]. The Cronbach’s alpha among this sample was 0.901.
Loneliness and Social Isolation Scale-6 items
The Loneliness and Social Isolation Scale-6 items (LSIS-6) scale is a self-report rating scale developed in English and Korean which can measure an individuals’ social isolation and loneliness [23]. It consists of 6 items rated on a 4-point scale, and a higher score reflects a higher level of loneliness. The Cronbach’s alpha among this sample was 0.747.
Statistical analysis
Demographic data were presented in means and standard deviations, or frequencies with proportions. We constructed linear regression models with the PGS, FCV-19S, LSIS-6, and PHQ-9 scores, each separately, and then altogether, as independent variables and the OCS score as the dependent variable. The relationships determined by the correlations were then used as a framework to construct the causal mediation models. We constructed four separate models. In three models, the direct effect of the PGS score on the OCS score was assessed, with either the FCV-19S, PHQ-9, or LSIS-6 scores as mediators. In the fourth model, the direct effect of the PGS score on the PHQ-9 score was analyzed, with the LSIS-6 score as the mediator. Bootstrapping with 1,000 iterations was used. Finally, based on the causal mediation analysis results, we constructed a structural equation model that included the PGS, FCV-19S, LSIS-6, PHQ-9, and OCS scores. Bootstrapping with 2,000 iterations was used for this analysis. We used R version 4.1.2 (R Core Team, www.r-project.org), with the dplyr, mediation and lavaan packages for analysis.
RESULTS
The participants were 267 healthcare workers who had experienced patient deaths during the past 2 years of the pandemic. About one third (34.1%) were doctors and the rest were registered nursing professionals (65.9%) (Table 1). They were aged 31 years on average and 78.7% were females. All participants had cared for COVID-19 patients, and 62 (23.2%) reported that the patients’ deaths were related to COVID-19.
Through linear regression analysis, we found that whereas the PGS, FCV-19S, LSIS-6, and PHQ-9 scores all showed some level of correlation with the OCS, only the associations of the FCV-19S and PHQ-9 with the OCS were independent of the others (estimate: 0.287, p<0.001 for the FCV-19S; estimate: 0.157, p<0.001 for the PHQ-9) (Table 2). The causal mediation analysis showed that the PGS scores had a significant indirect effect on the OCS scores, which was mediated by both the FCV-19S (regression coefficient: 0.324, p<0.001, proportion mediated: 0.51, 95% confidence interval [CI]: 0.37–0.66, p<0.001) and the PHQ-9 scores (regression coefficient: 0.498, p<0.001, proportion mediated: 0.77, 95% CI: 0.50–1.08, p<0.001) (Figure 1 and Table 3). These results show that viral anxiety and depression each mediated 51% and 77% of the effect of grief on obsession with the coronavirus when analyzed separately. Additionally, although LSIS-6 did not influence the OCS scores, it mediated the effect of the PGS scores on the PHQ-9 scores.
Based on these findings, we constructed a structural equation model for a comprehensive comparison of the variables (Figure 2). The FCV-19S and PHQ-9 scores both mediated the effects of the PGS scores on the OCS scores, and the strength of each was comparable at 29% and 22%, respectively (PGS → FCV-19S → OCS, regression coefficient: 0.286, p<0.001; PGS → PHQ-9 → OCS, regression coefficient: 0.220, p=0.003). Furthermore, the direct effect of the PGS scores on the OCS scores was not significant (regression coefficient: 0.138, p=0.128), reconfirming that most effects of the PGS scores on the OCS scores were not direct, but mediated by the PCV-19S and PHQ-9 scores.
DISCUSSION
We found that both viral anxiety and depression mediated the effect of grief on preoccupation with COVID-19 in healthcare workers who have recently experienced patient deaths. Loneliness was also a mediator between grief and depression. Furthermore, the indirect effects mediated by viral anxiety and depression were comparable to each other and accounted for most of the effects of grief on preoccupation with COVID-19.
The pandemic has been linked to various mental health problems in healthcare workers, and have caused issues with grieving [4]. Complicated grief has been linked to depression and anxiety [24], and a study reported that loneliness predicted both depression and anxiety symptoms during the pandemic [25]. Our results support these findings and offer a framework with which to approach their interactions, and show that grieving healthcare workers should be assessed for anxiety and depression and help to combat loneliness.
Our structural equation model can be interpreted with regards to the Terror Management Theory (TMT). The TMT stipulates that, when experiencing conscious thoughts of death, we engage in “proximal defenses,” and when those thoughts leave our consciousness, “distal defenses” are applied. The proximal defenses include thought suppression, denial of vulnerability, and measures to prevent death. As our participants had recently experienced the death of a patient, they would have been having conscious thoughts of death, which through anxiety and depression, increased obsessive thoughts and preoccupation with the coronavirus, and led to compulsive measures to prevent contracting COVID-19, and possible death. Although we did not assess such behavior in our participants, obsessive thoughts can lead to compulsive behavior [26], and obsessive-compulsive symptoms in healthcare workers have actually increased after the onset of the pandemic [27], supporting this argument.
This study has several limitations. First, a self-report questionnaire was used to assess depression. However, we used a widely used instrument; therefore, this probably had minimal effect [21]. Second, the age of our subjects was comparatively young. Young healthcare workers with limited work experience or those without proper training are more susceptible to depression or post-traumatic stress [5]. Therefore, the results of this study may be an overestimation of the average experience of a healthcare worker. Third, the participants were not assessed immediately after the death of a patient. Fourth, the participants were not limited to those who experienced COVID-related deaths only. The number of participants with COVID-related deaths was too small to have sufficient statistical power for our study objective. Larger studies will be needed to verify our results.
In conclusion, we found that viral anxiety and depression are associated with preoccupation with COVID-19 and mediate most of the effects of grief on preoccupation with COVID-19 in healthcare workers who have recently experienced the death of a patient. Furthermore, loneliness mediated the effect of grief on depression. Contingency plans for the next pandemic are being formulated by governments and institutions, and effective measures to protect the mental health of healthcare workers will be needed. Such measures should include assessments for grief, viral anxiety, depression, and loneliness.
Notes
Availability of Data and Material
The datasets generated or analyzed during the study are available from the corresponding author on reasonable request.
Conflicts of Interest
Seockhoon Chung, a contributing editor of the Psychiatry Investigation, was not involved in the editorial evaluation or decision to publish this article. All remaining authors have declared no conflicts of interest.
Author Contributions
Conceptualization: Seyoung Seo, Seockhoon Chung. Data curation: Eulah Cho. Formal analysis: Hoyoung An. Writing—original draft: Hoyoung An. Writing—review &editing: Hyejin Seo.
Funding Statement
None
Acknowledgements
None