Public servants have experienced an enormous workload since the outbreak of COVID-19. The longer the pandemic persists, the longer they will have to grapple with crippling work pressure. A study has confirmed that such workplace-related changes might affect public workers’ psychosocial abilities, which lead to burnouts, sick leave, and a decline in work performance and work motivation [8].

Perry and Wise [9] defined public service motivation (PSM) as ‘‘an individual’s predisposition to respond to motives grounded primarily or uniquely in public institutions and organizations.’’ A concern for the community and an eagerness to serve the common interest seem to be the major driving factors for public employees, which refers to a work ethic that values extrinsic rewards over intrinsic rewards [10]. It has three motives: the rational motive, the norm-based motive, and the affective motive [11]. The rational motive reflects the individual’s utility maximization, which enables them to engage in policymaking processes and defend special interests. The norm-based motive alludes to an aim for public interest and social equity and can be explained as patriotism and loyalty to the government [9]. The affective motive represents humanity, which refers to a volition to help others, and includes altruism, compassion (COM), and self-sacrifice (SS) [9].

Work engagement was first conceptualized by Kahn [12] as a state of total involvement in work-related roles physically and mentally. Later, it was categorized in three dimensions: vigor, dedication, and absorption [13]. Vigor is defined as high levels of energy and mental resilience during work time [14]. Dedication is strongly correlated with one’s volition to work, experiencing a feeling of significance, enthusiasm, inspiration, and pride [14,15]. Absorption is characterized as being in a state of full concentration and devoting oneself happily to one’s work, where time flies and it could become difficult to part from the endeavor [16,17].

Studies have repeatedly disputed the associations between PSM and work engagement as well as the affecting factors for each during non-pandemic years. Deng et al. [18] reported that PSM is highly influenced by stress among Chinese healthcare workers—positively by challenge stress and negatively by hindrance stress. Hindrance stress includes unnecessary impediments that prevent career development and goal achievement, such as organizational politics, job insecurity, and role conflict [19,20]. However, PSM has a positive effect on job performance [21], and each PSM dimension affects work engagement differently [22]. Additionally, it has been reported that work stress could directly or indirectly lower work efficiency by undermining PSM [23]. Resilience, which refers to one’s ability to endure stressful situations or cope with environmental adversity [24], was found to be a partial mediator between work stress and burnout symptoms among civil servants [25]. Work stress is, presumably, among the most significant antecedent factors that negatively affect work engagement, through the possible mediating parameters, PSM, or resilience during the pandemic era.

Psychological distress associated with COVID-19 had a strong link with burnout symptoms according to a study in Turkey [26]. Another study of public servants had confirmed resilience as an effective mediator and moderator when explaining the association between stress, anxiety, and depression [27]. Similarly, we have investigated the role of resilience, which would intervene the relationship between depression and anxiety reactions during the pandemic [28]. We verified a hypothesis that resilience could provide appropriate coping strategies for public servants to overcome mental stress during the pandemic. Anxiety and depressive symptoms have not been debated as predicting factors for PSM, and, to the best of our knowledge, no study has yet investigated the relation between PSM and pandemic-induced stress. Although PSM has to be considered differently from ordinary job motivation, stress related to the pandemic might impair PSM, eventually provoking depression.

This study investigated the influence of public workers’ work-related stress or viral anxiety on their depression and work engagement during the COVID-19 pandemic. Additionally, we explored whether their PSM or resilience mediates this association.

This online survey study was conducted with the help of a professional survey company (www.embrain.com). The respondents were notified of the study’s objective and enrollment procedure, and their participation was voluntary. The sample size was estimated based on the principles of 30 participants per cell [29]. We assigned 30 to 40 samples for each 10 cells, which were categorized by biological sex (two groups) and age (five groups). Approximately 5,000 to 6,000 public worker panelists received an email from the company for registration of the study. All 300 eligible participants’ responses were collected among 1,451 panelist who showed interest in undertaking the survey. This stands for about 0.03% of all registered public workers (1,068,626 based on the census of public workers in 2018) in South Korea [30].

From April 1 till April 12, 2021, 300 public workers were enrolled. The survey was conducted anonymously and no personal information was collected. The study protocol was approved by the Institutional Review Board (2021-0448) of the ASAN Medical Center, and the written informed consent requirement was waived. The survey form was developed according to the Checklist for Reporting Results of Internet e-Surveys guidelines [31], and an investigator (S.C.) tested the usability and technical functionality prior to implementation. We collected information about the participants’ age, sex, roles, years of employment, and marital status. Additionally, we included items related to COVID-19: “Did you experience being quarantined because of infection with COVID-19?” and “Did you experience being infected with COVID-19?” To ascertain their past and present psychiatric symptoms, we asked: “Have you experienced or undergone treatment for depression, anxiety, or insomnia?” and “Now, do you think you are depressed or anxious, or do you need help for your mood state?”

Statistical analysis was conducted using SPSS ver. 21.0, AMOS ver. 27 (for Windows; IBM Corp., Armonk, NY, USA), JASP ver. 0.14.1 (https://jasp-stats.org/), and Jamovi ver. 1.6.23 (https://www.jamovi.org). Clinical variables were summarized as mean±standard deviation, and the significance level was defined as two-tailed, p<0.05. To examine the expecting variables for public workers’ work engagement, the participants were categorized into two groups: UWES-9 top 25% group and UWES-9 bottom 75% group. Chi-square tests for categorical variables and the Student t-test for continuous variables were performed to explore the between-group difference. Pearson’s correlation analysis was performed to explore the correlations among clinical variables and rating scales scores. The linear regression analysis revealed the expecting variables for the high work engagement. Finally, the bootstrap method with 2,000 resamples was implemented to explore the mediation effect of resilience on work-related stress with work engagement.

It has already been reported that hindrance stress adversely impacts one’s psychological state, thus causing exhaustion and loss of passion and drive to work [42]. We found that high levels of work-related stress was significantly related with low levels of work engagement, consistent with studies conducted before the COVID-19 pandemic. Our results were similar to those of a study about the negative effects of stress and workload on work engagement of frontline nurses after the outbreak of COVID-19 [43].

High levels of work-related stress had a negative effect on work engagement and mediated partially in PSM and resilience. Based on the results of this study, which show the difference of the two groups categorized by the degree of work engagement, work-related stress, PSM, and resilience were more comparably distinctive. As revealed in previous studies, PSM tends to decrease under the influence of work stress and undermined PSM may lower work performance [23]. In this work, PSM was highly correlated with work-related stress and work engagement compared to other parameters according to Spearman’s correlation analysis. Both the linear regression analysis and the mediating analysis showed that PSM could partially explain the relation between work-related stress and work engagement of public workers. Resilience was closely correlated with work-related stress and work engagement, partially mediating the relation between them. This result corresponds with an earlier study [25] that suggested resilience as a mediating factor for explaining the relation between work stress and burnout during the non-pandemic era. There was no exact report regarding the relationship among the above-discussed factors for public workers during the pandemic. However, those relations were identically reproduced under the pandemic circumstances in our study, thereby suggesting that association between work-related stress and work engagement and mediating effect of PSM and resilience are not quite impacted by this extraordinary situation.

Although resilience and PSM were significant mediating factors, we should not overlook the fact that relationship between work-related stress and work engagement is mainly explained by the direct impact of work-related stress, rather than by partial mediating factors that reflect individual’s characteristics. Further research focusing on the comparison of strength of the direct impact of work-related stress on work engagement before and after the outbreak would be valuable to evaluate the actual effect of the pandemic on those relationships.

Year of employment was positively associated with work engagement; however, its value was not prominent, and it was not a significant factor to expect the level of work engagement. Longer work periods would reflect their improved ability in work performance; conversely, physical or mental fatigue from long-lasting work could decrease the concentration on the task. These divergent explanations might have resulted in a weak association between the two factors.

Viral anxiety did not have a significant effect on public workers’ work-related stress and work engagement, which was contrary to our expectation. South Korea has been recognized as an example to emulate in terms of promptly established preventive measures against the pandemic [44]. Moreover, the survey was conducted a year after the outbreak, by which time systemic changes had already been settled. Therefore, anxiety reactions might not have had actual effects on public workers’ performance. However, it is noteworthy that participants’ scores on the SAVE-6 scale were comparably higher than in previous reports of other occupational groups. While acknowledging that a simple comparison may be inappropriate in the following case, we share that the average score on the SAVE-6 scale for healthcare workers at two big hospitals in South Korea was 14.4±4.5, whereas the average scores in case of the general South Korean population were 14.4±4.5 (male) and 14.7±4.6 (female) [32], which were lower than our results. If public workers’ viral anxiety was substantially higher than other groups, consequent changes from anxiety to work engagement might not be apparent. Considering that viral anxiety was related with resilience and depression, which could further influence work engagement, it would be difficult to negate the possibility that viral anxiety might be a factor affecting work engagement.

Pandemic-induced psychological distress was reportedly closely linked with mood symptoms [45,46]. According to a study about the mental health status of university students in Bangladesh, depressive symptoms were significantly expected by the perception of pandemic state and the concern for the severity of COVID-19 [47]. In this study, we observed that viral anxiety significantly influenced depressive symptoms, and resilience mediated this association. We had previously reported similar results among other samples of public workers [28]. Similar results were also reported among schoolteachers [48]. Additionally, it has been noted that public servants’ anxiety is highly correlated with depression and resilience mediated the association between stress, anxiety, and depression during the nonpandemic years [27]. Consequently, anxiety reaction, regardless of whether it was derived from the viral pandemic or not, contributes to depression and resilience acts as a mediator between them in case of public workers.

Resilience was strongly correlated with depression compared with other factors according to Pearson’s correlation analysis. Further, the effect of resilience on depression was more prominent in the mediating analysis, while the direct effect of viral anxiety was also significant. Though viral anxiety might lower the individual capacity to endure stress state, which eventually decreases the susceptibility to depression, resilience itself independently exerts its power to reduce the probability of having depressive symptoms, which accords to previous reports [49,50]. Thus, caution should be exercised while interpreting the results, as the role of personal vulnerability is also relevant.

Year of employment was negatively associated with depression, but the degree of the association was small, and it was not a meaningful factor to determine the level of work engagement. Work period was not related to viral anxiety and resilience, which were crucial underlying factors to determine the level of depression in our research. The individual capacity to endure stress and the lower level of anxiety might help one to work for longer periods, whereas other external factors such as a workload, relationships between coworkers, and organizational culture would exert negative influence on it. Moreover, despite the long career, an unfamiliar challenging state like the viral pandemic would be a difficult situation to endure. The year of employment is complicatedly determined through one’s internal and external aspects; therefore, there is no single factor to explain the susceptibility to depression.

In this study, public workers’ PSM did not mediate the association between their viral anxiety and depression. PSM was negatively correlated with depression, but the correlation coefficient was not substantial. Further, PSM was not a significant factor to determine the level of depression although it had a negative relationship with depression. It has several different interpretations. First, there was an insignificant positive correlation between PSM and viral anxiety. We can speculate that as public workers’ viral anxiety increases because of aggravated social conditions under the pandemic, motivation to resolve the situation may also rise, because their duties primarily aim to promote public interest, not personal gain. A recent study explained that emotional disturbance due to a threatening situation, such as a pandemic, could provide the motivation to act [51]. However, considering earlier findings that anxiety has a negative relationship with motivation [52], it would be reproduced in this situation where an emotionally unstable state along with viral anxiety might depress the workers’ morale. These bidirectional approaches would form a nonexplicit conclusion that public workers’ PSM was not significantly affected by viral anxiety. Second, a comparatively higher level of PSM would be expected during the pandemic, if the initial response of the government was successful enough for the public to have confidence in the administration. As revealed in a study, public workers in China were highly motivated during the COVID-19 pandemic and their motivation had a positive correlation with public satisfaction and cooperation [53]. In our research, the PSM measured one year after the outbreak would be higher than the usual state, considering that Korea was regarded as a model of credible national quarantine system during that period. In case public workers have already been encouraged by external factors during the pandemic, consequent changes in PSM due to viral anxiety might not be apparent.

This study has several limitations. First, the survey was conducted in April 2021—14 months after the pandemic’s outbreak, which might influence the results. Public workers might have already adjusted to the pandemic situation, and that might be the reason for the lack of mediation effect of PMS on the relationship between viral anxiety and depression. Moreover, the Korean government announced the “living with COVID-19” policy on November 1, 2021, as the country logged about 7,000 daily confirmed cases [54]. The results can, therefore, vary depending on the severity of the pandemic situation. Second, we can speculate that the viral anxiety perceived by public workers can vary depending on their work role. Certain public workers might be in roles directly related with COVID-19, such as developing and executing the prevention policy, visiting or transferring infected persons, or cleaning and disinfecting. Similarly, others may be in roles not directly related with COVID-19. The results of this study, therefore, must be confirmed among public workers whose roles are directly related with COVID-19 in a further investigation. Third, this research was conducted via an anonymous online survey instead of face-to-face interviews to prevent the risk of a viral outbreak. However, the online survey might lead to a bias in the study design. Fourth, the relatively small sample size of the participants in this study can reduce its statistical power.

In conclusion, we observed that resilience and PSM partially mediate the effects of work-related stress on work engagement. Depression was influenced by COVID-19-induced viral anxiety, and resilience mediated this association. Especially in the relationship between viral anxiety and depression, resilience had a noticeable impact to lower the levels of depressed mood, which suggests that as a personal aspect, it might be one of the decisive factors to assess the susceptibility of depression. The results of this study might contribute to the development of a psychological support system for public workers in this pandemic era.