INTRODUCTION

Psychological well-being among frontline nurses is critical for delivering quality patient care during the COVID-19 Pandemic, declared by the World Health Organization (WHO) in March 2020. They faced a high risk of occupational SARS-CoV-2 infection as a part of their daily duties. At the beginning of the pandemic, the absence of approved drugs or vaccines, shortage of personal protective equipment (PPE), and lack of understanding of viral transmission have resulted in poor mental health among nurses (Vindegaard & Benros, 2020; Wang et al., 2020). For example, the proportions of nurses in the United States with moderate/severe anxiety and moderate/severe depression were almost five times higher in the early pandemic period than those in the pre-pandemic period (Kim, Quiban, et al., 2021). According to systematic reviews, nurses experienced even more psychological distress than physicians during the pandemic (Batra et al., 2020; Pappa et al., 2020). When there was a great need for healthy frontline nurses who could provide high-quality patient care, they had to bear the enormous psychological burden.

In July 2020, the first effective COVID-19 drug was reported. Daily dexamethasone showed a 36% reduction in mortality among hospitalized patients on mechanical ventilation.

(RECOVERY Collaborative Group, 2021). Subsequently, early outpatient treatment of patients with mild or moderate COVID-19 with a cocktail of SARS-CoV-2 neutralizing monoclonal antibodies showed a significant reduction in the risk of hospitalization (Weinreich et al., 2021). In addition to these therapeutic agents in the United States, two mRNA vaccines became available December 2020 under Emergency Use Authorization (EUA) with 94%–95% efficacy in preventing symptomatic COVID-19 (Baden et al., 2021; Polack et al., 2020). As a high-risk group for infection, the frontline healthcare workers received the first available vaccines in the United States ahead of other populations (Jean-Jacques & Bauchner, 2021). Studies among vaccinated healthcare workers showed significant reductions in symptomatic and asymptomatic COVID-19 infection (Angel et al., 2021; Benenson et al., 2021). Although the deployment of highly effective COVID-19 vaccines has improved the physical health of the frontline nurses, the influence of the vaccine on their mental health has not been thoroughly examined.

Despite various recommendations for resilience building to prevent poor mental health for the frontline healthcare workers during the pandemic, the evidence supporting such interventions is weak. For example, a Cochrane systematic review of 16 interventional studies performed during or after epidemics showed little evidence of improvements in resilience and mental health for the frontline healthcare workers (Pollock et al., 2020). Another review also noted that evidence for resilience building is weak. Still, the review highlighted the protective factors of family and social supports, physical activity, and spirituality associated with psychological resilience during the COVID-19 pandemic (Blanc et al., 2021). Previous observational studies have identified predictors related to nurses’ poor mental health during the COVID-19 pandemic. For example, high workload and poor family relationships were associated with 1.5-fold higher odds of depression (Zheng et al., 2021). In contrast, high levels of resilience, family functioning, and spiritual support were associated with two- to five-fold lower odds of stress, anxiety, and depression among nurses during the early COVID-19 pandemic period (Kim, Quiban, et al., 2021). In addition, organizational support, such as providing counseling services and stress-management training, and meeting the basic needs of the staff were protective factors for nurses’ mental health (Cao et al., 2020; Zhu et al., 2020).

We had previously reported nurses’ mental health status and coping mechanisms during the early pandemic period following the WHO’s COVID-19 Pandemic declaration as well as the period before the vaccines became available (Kim et al., 2021). However, changes in nurses’ mental health after the vaccines became available have not been thoroughly examined. The primary aim of this study was to explore the changes in nurses’ mental health over a 1-year time span, from the early pandemic to the early vaccination period. The secondary aim was to examine vaccination and coping mechanisms as predictors of nurses’ poor mental health and burnout following the vaccine availability. Lazarus and Folkman's Transactional Model of Stress and Coping (1984) provided the theoretical framework of how an individual's appraisal of stressor and available coping resources impacts the mental health. In our study, the coping resources included resilience, family support, and spiritual support.

METHODS Design and sample

Repeated cross-sectional surveys were conducted to collect data at three-time points during the pandemic via an online platform, QualtricsXM. Figure 1 shows the pandemic timeline and the study cohorts. The first cross-sectional survey of 320 nurses, defined as the Early-Pandemic Cohort, was conducted between April 20 and May 10, 2020 (Kim, Quiban, et al., 2021). A second cross-sectional survey was conducted among multidisciplinary healthcare workers from September 8 to October 10, 2020 (Kim, Sloan, et al., 2021). From this second survey, data for all nurses (n = 228) were extracted into a cohort defined as the Pre-Vaccination (Pre-Vax) Cohort. Finally, a third cross-sectional survey was conducted between March 4 and April 18, 2021, defined as Early-Vaccination (Early-Vax) Cohort. For the third survey, alumni who graduated from two nursing schools in southern California and Midwest, USA, were invited to participate in the study by emails containing the hyperlink to the online survey, followed by reminder emails sent two weeks later. Nurses who were inactive or retired from the nursing profession were excluded from the study. The sampling and data collection methods for the first two surveys have been reported previously (Kim, Quiban, et al., 2021; Kim, Sloan, et al., 2021).

Pandemic timeline and study cohorts (N = 840)

Note. The top graph shows daily SARS-CoV-2 infections and seven-day moving average curve in the United States (Coronavirus-Worldometer. https://www.worldometers.info/coronavirus/country/us/)

Survey

The survey consisted of various valid and reliable instruments to assess nurses’ mental health and coping mechanisms. It also included demographic information, quarantine/self-isolation experience, COVID-19 patient care, and satisfaction with the organization's COVID-19 procedures.

The 10-item Perceived Stress Scale (PSS) asks the participants’ perceptions about their feelings and thoughts during the past month on a 5-point Likert scale, ranging from 0 (never) to 4 (very often) (Cohen et al., 1983). The summation scores from 0–13, 14–26, and 27–40 indicate low, moderate, and high stress, respectively. Factorial validity was established and Cronbach's alpha was previously reported as 0.83 (Leung et al., 2010). In our study, Cronbach's alpha was 0.87 for the Early-Vax Cohort.

General Anxiety Disorder-7 (GAD-7) and Patient Health Questionnaire-9 (PHQ-9) were used to assess anxiety and depressive symptoms, respectively (Spitzer et al., 1999, 2006). Participants rated how often they have been bothered with the respective symptoms over the past two weeks on a 4-point Likert scale, ranging from 0 (not at all) to 3 (nearly every day). Summation scores of ≥10 indicate moderate/severe anxiety or moderate/severe depression. The sensitivity and specificity of GAD-7 were 89% and 82% for anxiety at a cutoff score of ≥10. Similarly, PHQ-9 had sensitivity and specificity of 88% for depression at a cutoff score of ≥10.

A global item was used by asking participants to rate the burnout severity based on their definition of burnout on a 5-point ordinal scale ranging from 1 (no symptoms of burnout) to 5 (feel completely burnout) (Dolan et al., 2015; Rohland et al., 2004). The psychometric validation of this global item against the Maslach Burnout Inventory-Emotional Exhaustion (MBI-EE) tool among 5404 healthcare workers showed the inter-rater reliability with Cohen's kappa of 0.70, the sensitivity of 83.2%, and specificity of 87.4% (Rohland et al., 2004).

The 10-item Connor-Davidson Resilience Scale (CD-RISC) assesses the ability to adapt to changes and cope with challenges over the past month on a 5-point Likert scale ranging from 0 (not true at all) to 4 (true all the time) (Connor & Davidson, 2003). With summation scores ranging from 0 to 40, higher scores indicate higher levels of resilience. Convergent validity and internal consistency reliability as Cronbach's alpha of 0.89 were reported in the previous study. In our study, Cronbach's alpha was 0.86 for the Early-Vax Cohort.

The 5-item Family APGAR tool assesses participants’ satisfaction in five areas of family functioning, such as Adaptation, Partnership, Growth, Affection, and Resolve, on a 3-point Likert scale ranging from 0 (hardly ever) to 2 (almost always) (Smilkstein et al., 1982). Higher summation scores indicate higher satisfaction with family function. The inter-item correlation coefficients of the internal consistency ranged from 0.63 to 0.71 (Gardner et al., 2001).

The 12-item Spirituality Support Scale assesses the participants’ faith experience or perceived support from higher powers on a 4-point Likert scale, ranging from 1 (strongly disagree) to 4 (strongly agree) (Ai et al., 2005). The summation scores range from 12 to 48, and a higher score indicates higher spirituality support. Construct validity and Cronbach's alpha of 0.97 were reported in the previous study. Cronbach's alpha in our study was 0.98 for the Early-Vax Cohort.

Data analysis

Means, standard deviations, frequencies, and percentages were calculated. Poor mental health was defined as moderate/high stress (PSS scores ≥14), moderate/severe anxiety (GAD-7 scores ≥10), and moderate/severe depression (PHQ-9 scores ≥10). Severe burnout was defined as the responses of 4 or 5. The coping mechanisms were recoded as dichotomous variables as described previously (cutoff scores of 30, 39, and 10 for resilience, spirituality, and family functioning, respectively) (Kim,Quiban, et al., 2021; Kim, Sloan, et al., 2021). COVID-19 vaccination was also recoded as dichotomous variable (full vaccination with two doses = 1 vs. partial vaccination/no vaccination = 0).

Pearson's chi-square tests were used to compare the proportion of nurses with poor mental health among the three cohorts. For cohorts with significant differences, post hoc analyses with Bonferroni adjustment for multiple testing were carried out. For the Early-Vax Cohort, Kendall's tau's bivariate correlation procedures were performed among poor mental health, COVID-19 vaccination status, various coping mechanisms, and demographic variables. All statistically significant variables from the Kendall's tau tests were entered into multivariate logistic regression procedures to examine the predictors of poor mental health. Data were analyzed using the SPSS version 26.0 (IBM Corporation), and the significance level was set at p-value <0.05.

Ethical considerations

The study of the Early-Vax Cohort was reviewed and approved by the Institutional Review Boards of the universities. A waiver of documented informed consents was granted due to the minimal risks involved in this online survey study, and completion of the survey indicated the consent to participate. The respondents were reminded that their participation was entirely voluntary and their choice to participate would not affect their relationship with the universities.

RESULTS Sample characteristics

The sample characteristics of the three cohorts are shown in Table 1: Early-Pandemic, Pre-Vax, and Early-Vax Cohorts. They have similar characteristics except for the higher proportion of nurses with COVID-19 patient care over time (48.8% vs. 60.5% vs. 70.2%, respectively) and COVID-19 vaccination for the Early-Vax Cohort only (71.9%). The vaccines were not available for the first two cohorts.

TABLE 1. Sample characteristics (N = 840) Early pandemic Cohort (n = 320) Pre-Vax Cohort (n = 228) Early-Vax Cohort (n = 292) Age 33 (21–67) 38 (22–68) 40 (21–73) Gender Female 302 (94.4) 201 (88.2) 265 (90.8) Male 18 (5.6) 27 (11.8) 27 (9.2) Ethnicity White 234 (73.1) 168 (73.7) 214 (73.3) African Americans 5 (1.6) 12 (5.3) 21 (7.2) Hispanics 33 (10.3) 10 (4.4) 20 (6.8) Asian/Pacific Islanders 33 (10.3) 24 (10.5) 20 (6.8) Others 15 (4.7) 14 (6.1) 17 (5.9) COVID-19 patient care 158 (48.8) 138 (60.5) 205 (70.2) Quarantine/self-isolation experience 147 (45.4) 88 (38.6) 155 (53.1) COVID-19 vaccinationa VNA VNA 210 (71.9) Note Data for the early-pandemic are from previously published paper (Kim et al., 2021). Values are expressed as n (%) unless otherwise indicated. The percentage may not add to 100% because of missing data or rounding. Abbreviation: VNA, Vaccine Not Available. Changes in nurses’ mental health over a 1-year time span of the pandemic

Figure 2 depicts the percentage of nurses with moderate/high stress (80% vs. 81% vs. 74%), moderate/severe anxiety (43% vs. 37% vs. 28%), and moderate/severe depression (26% vs. 32% vs. 20%) for each cohort, respectively. For the Pre-Vax and Early-Vax Cohorts, the nurses with severe burnout were 30% and 14%, respectively. There was a significant association between the three cohorts and moderate/severe anxiety (χ2(2) = 13.1; p = 0.001) with a significant decrease from Early-Pandemic Cohort to Early-Vax Cohort (43% vs. 28%). Likewise, a significant association was found between the three cohorts and moderate/severe depression (χ2(2) = 9.36; p = 0.009) with a significant decrease from Pre-Vax Cohort to Early-Vax Cohort (32% vs. 20%). For the severe burnout, there was a significant decrease from Pre-Vax Cohort to Early-Vax Cohort (30% vs. 14%; χ2(1) = 20.1; p < 0.001). However, no significant association was found between the three cohorts and moderate/high stress (χ2(2) = 4.16; p = 0.125).

Poor mental health in Early-Pandemic, Pre-Vax, and Early-Vax Cohorts (N = 840)

Note. White bars, crosshatched bars, and black bars represent the percentages of nurses in Early-Pandemic, Pre-Vax, and Early-Vax Cohorts, respectively. Statistically significant associations were found between cohorts and moderate/severe anxiety (p = 0.001), moderate/severe depression (p = 0.009), and severe burnout (p < 0.001) by Pearson’s chi-square tests. †Burnout was not collected for the Early-Pandemic Cohort

Predictors of poor mental health and burnout in Early-Vax Cohort

The results of bivariate Kendall's Tau correlations are shown in Table 2. COVID-19 vaccination was correlated only with moderate/severe anxiety but in a positive direction (r = 0.19; p = 0.004). Caring for COVID-19 patients or quarantine/self-isolation experiences also correlated positively with moderate/severe anxiety, moderate/severe depression, and severe burnout. In contrast, satisfaction with the organization's COVID-19 procedures, including PPE availability, communication, or infection control processes, had negative correlations with all measures of poor mental health. Likewise, there were weak negative correlations between various coping mechanisms and poor mental health. Among demographic variables, only age was found to have negative correlations with poor mental health and severe burnout.

TABLE 2. Correlations with poor mental health and burnout in Early-Vax Cohort (N = 292) Moderate/high stress Moderate/severe anxiety Moderate/severe depression Severe burnout COVID-19 vaccinationa −0.09 0.19** −0.01 0.06 Age −0.12* −0.19** −0.20*** −0.11* COVID-19 patient care 0.01 0.14* 0.16* 0.014* Quarantine or self-isolation experience −0.06 0.19** 0.17** 0.06 Feel protected working with COVID-19 patients −0.10 −0.05 −0.17* −0.12 Satisfaction with PPE availability −0.12 −0.07 −0.16* −0.23*** Satisfaction with organization's infection control processes −0.15* −0.19** −0.25*** −0.29*** Satisfaction with organization's COVID-19 communication −0.04 −0.17* −0.18** −0.28*** High resilience −0.33*** −0.09 −0.13* −0.11 High family functioning −0.25*** −0.14* −0.22** −0.19** High spirituality −0.19** −0.20** −0.18** −0.12 Note Correlation coefficients by Kendall's Tau test.

Table 3 shows the results of the multivariate logistic regression procedures. COVID-19 vaccination or quarantine/self-isolation experience, respectively, was associated with almost three-fold (OR = 2.87; 95% CI = 1.28–6.44) or two-fold (OR = 2.00; 95% CI = 1.06–3.76) higher odds of moderate/severe anxiety. In contrast, satisfaction with the organization's COVID-19 communication was associated with five-fold lower odds of severe burnout (OR = 0.20; 95% CI = 0.09–0.43) and two-fold lower odds of moderate/ severe anxiety (OR = 0.47; 95% CI = 0.25–0.90), while satisfaction with the organization's infection control processes was associated with three-fold lower odds of moderate/severe depression (OR = 0.30; 95% CI = 0.14–0.60). Similarily, many of the coping mechanisms were negative predictors of poor mental health and severe burnout. For example, high family functioning was associated with two- to three-fold lower odds of moderate/high stress (OR = 0.44; 95% CI = 0.23–0.84), moderate/severe depression (OR = 0.32; 95% CI = 0.15–0.71), and severe burnout (OR = 0.29; 95% CI = 0.12–0.70). High spirituality was associated with nearly two-fold lower odds of moderate/high stress (OR = 0.52; 95% CI = 0.27–0.98), whereas high resilience was associated with four-fold lower odds of moderate/high stress (OR = 0.24; 95% CI = 0.12–0.48). Among demographic variables, every 1 year of age was associated with 3% lower odds of moderate/severe anxiety (OR = 0.97; 95% CI = 0.94–0.99) and 5% lower odds of moderate/severe depression (OR = 0.95; 95% CI = 0.92–0.99). In contrast, age was not a significant predictor of severe burnout (OR = 0.97; 95% CI = 0.93–1.00; p = 0.054).

TABLE 3. Predictors of poor mental health and burnout in Early-Vax Cohort (N = 292) OR 95% CI p-value Moderate/high stress High resilience 0.24 0.12–0.48 <0.001 High family functioning 0.44 0.23–0.84 0.013 High spirituality 0.52 0.27–0.98 0.042 Moderate/severe anxiety Age 0.97 0.94–0.99 0.009 COVID-19 vaccinationa 2.87 1.28–6.44 0.010 Quarantine or self-isolation experience 2.00 1.06–3.76 0.032 Satisfaction with organization's COVID-19 communications 0.47 0.25–0.90 0.024 Moderate/severe depression Age 0.95 0.92–0.99 0.005 Quarantine or self-isolation experience 2.73 1.26–5.91 0.011 High family functioning 0.32 0.15–0.71 0.005 Satisfaction with organization's infection control processes 0.30 0.14–0.60 0.001 Severe burnout Age 0.97 0.93–1.00 0.054 Satisfaction with organization's COVID-19 communications 0.20 0.09–0.43 <0.001 High family functioning 0.29 0.12–0.70 0.006 Abbreviations: OR, Odds ratio; CI, Confidence Interval DISCUSSION

To our knowledge, this is the first report of changes in frontline nurses’ mental health over 1 year, from the early pandemic to the early vaccination period, as well as the influence of the COVID-19 vaccination on their mental health. The study findings indicate that nurses are still experiencing poor mental health 1 year after the pandemic declaration and vaccine availability. Remarkably, the proportion of nurses with moderate/high stress changed little over the year. In contrast, a smaller proportion of nurses had moderate/severe anxiety, moderate/severe depression, or severe burnout after 1 year. Severe burnout showed the most significant decrease from the pre-vaccination period (30%) to the early-vaccination period (14%). Perhaps, the availability of preventive and therapeutic agents, as well as a better understanding of the disease, may have reduced poor mental health and burnout.

It was surprising that the vaccinated nurses in Early-Vax Cohort were more likely to have moderate/severe anxiety than the unvaccinated nurses, as shown by bivariate and multivariate analyses. Because this finding from our observational study does not indicate cause and effect relationship, it would be erroneous to conclude that vaccination causes anxiety. Instead, we believe that anxious nurses are more likely to get vaccinated. This is consistent with studies indicating that healthcare workers with higher anxiety or fear of COVID-19 are more willing to get vaccinated (Sun et al., 2021; Szmyd et al., 2021).

It was found that the nurse's age was a significant negative predictor of poor mental health, with each year of age being associated with 3%–5% lower odds of anxiety and depression. Similar findings have been reported previously that older nurses experience lower levels of poor mental health and burnout during the COVID-19 pandemic (Zhang et al., 2020). This might be due to the older nurses’ having a wealth of life experience and being more confident in healthcare crises (Raveis et al., 2020). Therefore, it may be helpful for the older, seasoned nurses to mentor younger nurses and provide support during the pandemic.

We have examined the influence of various coping mechanisms of family support, resilience, and spirituality on nurses’ mental health. Among these, family functioning appears to be the broadest protective factor against multiple dimensions of poor mental health. Others have also found similar results during the pandemic (Alnazly et al., 2021; Tselebis et al., 2020). The government-mandated social distancing and shutdowns likely heightened the importance of close family relationships during the pandemic. In addition, high spirituality and resilience were associated with a two- to four-fold decrease in moderate/high stress among nurses in the early vaccination period. Although resilience has been widely recognized as an important coping resource, there is an urgent need for high-quality evidence supporting resilience-building interventions to help frontline nurses cope with major crises like the pandemic. Instead of focusing on a narrow resilience-building activity, it may be helpful to consider multidimensional approaches, including family, social and spiritual supports, as well as physical exercise and individualized counseling (Blanc et al., 2021). In addition, intervention enhancing the internal motivation to the nursing profession may be helpful for resilience. It was found that healthcare professionals with a sense of higher calling and commitment to their works during the pandemic reported lower levels of anxiety, depression, and burnout (Hartzband & Groopman, 2020; Raveis et al., 2020).

Satisfaction with the organizational supports, such as COVID-19 communications or infection control processes, was associated with lower levels of poor mental health and burnout in bivariate and multivariate analyses in this study. During the pandemic, many organizations’ leadership developed and provided pandemic planning roadmaps to their employees and communicated them clearly, which may have contributed to their good mental health (Hardt DiCuccio et al., 2020).

Limitations

There are several limitations to this study. First, the associations between COVID-19 vaccination or coping mechanisms with poor mental health should not be considered cause-and-effect relationships in this observational study. Second, although this study reported changes in nurses’ mental health over 1 year, from the early pandemic to the early vaccination period, it was not a longitudinal study following the same subjects. In addition, the pandemic was also changing rapidly with intermittent surges of infections, which may have caused unmeasured changes in the subjects’ work environment. Nevertheless, all three cohorts had similar measured characteristics except for the vaccination and increasing COVID-19 patient care over time. Third, convenience sampling with self-selection and the subjective rating of their mental health may have introduced bias. Finally, this study was conducted in southern California and Midwest, USA. Thus, the study finding may not be generalizable to other countries where the vaccination rollout pattern and nature of the available vaccines may differ. Future interventional studies are urgently needed to evaluate the impacts of vaccination, various coping mechanisms, and organizational support on nurses’ mental health during and after the pandemic.

CONCLUSION

This study reports nurses’ changes in mental health status over 1 year from the beginning of the COVID-19 pandemic to the early vaccination period. Although the proportion of nurses with anxiety, depression, and burnout decreased over 1 year, the COVID-19 vaccination seemed to have minimal associations with nurses’ mental health except for anxiety. It appears that coping mechanisms and organizational support are more important predictors of nurses’ mental health than vaccination status. The evidence gathered over 1 year of the pandemic may be helpful for a better understanding of the challenges facing frontline nurses caring for the sick and preparing for future healthcare crises. As a part of the preparedness plan for the future, evidence-based interventions that raise nurses’ resilience, as well as family and spiritual support, should be considered.