Table 1.
General study characteristics.
| Author, year, country of origin | Study aim | Study design | Participants (n) |
Key findings | Limitations | Quality appraisal (JBI score) |
|---|---|---|---|---|---|---|
| Altinbilek et al. (2021) Turkey [31] |
To investigate the stress and anxiety of healthcare workers due to COVID-19 | Prospective multi-centre survey in two hospitals | 205 ED healthcare workers | Physicians and nurses had higher anxiety and stress scores than other staff, and scores of nurses were significantly higher than physicians (p < 0.05). | Survey was conducted online for only one week with no evidence of reminder. | 10/12 |
| An et al. (2020) China [23] |
To examine the prevalence of depressive symptoms (depression hereafter) and their correlates and association between depression and quality of life | National, cross-sectional online survey | 1103 frontline ED nurses | Overall prevalence of depression in nurses was 43.61% (95% CI = 40.68–46.54%). Multiple logistic regression analysis revealed that working in tertiary hospitals (OR = 1.647, p = 0.009), direct patient care of COVID-19 patients (OR = 1.421, p = 0.018), and current smokers (OR = 3.843, p < 0.001) were significantly associated with depression. | Some variables associated with depression, such as social support, collegial relationship, health status and pre-existing psychiatric disorders, were not examined. Because of the cross-sectional study design, causality between depression and other variables could not be examined. | 8/8 |
| More than 90% of participants were female nurses, which may have biased the findings. | ||||||
| Araç and Dönmezdil (2020) Turkey [28] | To examine psychiatric disorders such as anxiety, depression and sleep disorders among healthcare professionals working in an emergency department and a COVID-19 clinic | Cross-sectional survey | 198 healthcare professionals: 100 in ED and 98 in a COVID-19 clinic | Perceived stress levels and PSQI sub-scores were found to be significantly higher among volunteers working in the emergency department than among those in other departments. The risk of development of anxiety among women was 16.6 times higher than among men. | The study was conducted at a single academic medical centre at one point in time and was limited to a relatively small sample size and low response rate. The authors also identified there were limitations of the tests used in the study. | 8/8 |
| Gender (OR = 16.631, p 0.001), profession (physicians, OR = 8.750, p = 0.022; nurses, OR = 4.845, p = 0.045), HADS-depression (OR = 9.194, p = 0.002), the use of sleeping medication subscale score of the PSQI (OR = 6.357, p = 0.012), and the perceived stress level (OR = 8.639, p = 0.003) were found to be effective risk factors of mental health. | ||||||
| Baumann et al. (2021) USA [21] | To provide a longitudinal assessment of anxiety levels and work and home concerns of U.S. emergency physicians during the COVID-19 pandemic | Longitudinal, cross-sectional email survey | 262 physicians | In examining the relationship between demographics, living situations, and institution location on having a PC-PTSD-5 score ≥ 3, only female sex was associated with a PC-PTSD-5 score ≥ 3 (adjusted OR = 2.48, 95% CI = 1.28–4.79). | The researchers reported a low response rate (61.5%) and no sample size calculation was evident. | 8/8 |
| There were different phases of the pandemic at various sites during the study period so responses at the time of reporting may vary. | ||||||
| A lack of baseline stress, burnout and PTSD measures from before the pandemic is problematic. | ||||||
| The authors noted a discrepancy between median burnout stress scores and PC-PTSD-5 screener ≥ 3 scores. | ||||||
| Caliskan and Dost (2020) Turkey [29] | To evaluate emergency physicians’ levels of depression and anxiety in dealing with the COVID-19 pandemic | Descriptive, cross-sectional study using an online questionnaire | 290 emergency physicians | According to the physicians’ HADS scores, depression was detected in 180 participants (62%) while anxiety was detected in 103 (35.5%), with median depression and anxiety scores found to be 8 (0–21) and 7 (0–21), respectively. | The data and relevant analyses presented were derived from a cross-sectional study design. Thus, it was difficult to make causal inferences. Secondly, the authors used a web-based survey method to avoid bias, necessitating inclusion of volunteer participants; therefore, the possibility of selection bias should be considered. | 7/8 |
| Cui et al. (2020) China [24] | To identify the impact of COVID-19 on the psychology of Chinese nurses in emergency departments and fever clinics and to identify associated factors | Online cross-sectional study | 453 nurses in ED and fever clinics | Fear of infecting family members was the most influential and predictive of anxiety (β = 0.263, p = 0.000), stress (β = 0.239, p = 0.000), and inverse stress coping tendency (β = − 0.188, p = 0.000) of nurses. | The method of snowball sampling may reduce generalisability of the results. Secondly, the data’s cross-sectional nature constrained making causal links among anxiety, stress, and coping tendency. Finally, the survey relied only on self-reported questionnaires, which may reduce data collection objectivity. | 7/8 |
| de Wit et al. (2020) Canada [39] | To report burnout time trends and describe psychological effects of working as a Canadian emergency physician during the first weeks of the coronavirus disease 2019 (COVID-19) pandemic | Mixed-methods study (both quantitative and qualitative) | 468 physicians | Being tested for COVID-19 (OR = 11.5, 95% CI = 3.1–42.5) and the number of shifts worked (OR = 1.3, 95% CI = 1.1–1.5 per additional shift) were associated with high emotional exhaustion. Having been tested for COVID-19 (OR = 4.3, 95% CI = 1.1–17.8) was also associated with high depersonalisation. | The authors identified possible reasons why the study did not establish association between burnout and progression of time during the first weeks of the pandemic including: | 7/10 & 8/8 |
| (1) psychological distress was associated with high levels of exposure to traumatic events, which may explain the lack of worsening burnout levels over time during the pandemic where the cohort possibly did not experience rigorous numbers of sick patients seen in other parts of the world. | ||||||
| (2) mental health effects of the pandemic, particularly burnout, may have been measured too soon after the start of the pandemic. | ||||||
| Elhadi et al. (2021) Libya [34] | To examine prevalence of anxiety, depression, and burnout among frontline emergency physicians of the COVID-19 pandemic | Cross-sectional survey | 154 physicians | Comparison between groups of physicians experiencing anxiety (HADS anxiety ≥ 11) and depression (HADS depression ≥ 11) demonstrated the following elements to be statistically significant: for anxiety, only age range, working hours per week, and transport issues were associated with a higher prevalence of anxiety (p < 0.05). | Firstly, the study focused only on one country, where several additional and unique factors may have contributed to high levels of mental distress. These aspects may, in turn, have aggravated COVID-19 effects. Furthermore, interviewer bias may be present, as some respondents may have opted to hide or alter their responses out of a fear of stigmatisation, despite the anonymous nature of the survey. Additionally, due to the cross-sectional study design, lower causation and linkage ability may be apparent. | 7/8 |
| Fitzpatrick et al. (2020) USA [19] | To determine the effect that the COVID-19 pandemic had on the wellness of emergency physicians | Prospective survey | 55 physicians | During the pandemic, emergency physicians felt less in control (p = 0.001); felt decreased happiness while at work (p = 0.001); had more trouble falling asleep (p = 0.001); had increased sense of dread when thinking of work needing to be done (p = 0.04); felt more stress on days not at work (p < 0.0001) and were more concerned about their own health (p < 0.0001). | The sample size (n = 55) was relatively small. The population originated from a single hospital network, was a convenience sample, and was limited by non-response bias. Survey questions were derived from a previously validated study, but the specific question that subjects answered might not have covered the broad range of physician wellness. The survey used physician self-report of feelings up to six months earlier, which introduced the potential for recall bias, as well as social-desirability bias. Even though statistical significance was found in several questions, there may not be clinical significance given how similar the medians and/or general distribution of scores were in some cases. | 7/8 |
| Havlioğlu and Demir (2020) Turkey [30] | To determine anxiety levels of emergency service employees working during the COVID-19 pandemic | Descriptive, cross-sectional study | 95 emergency service employees | It was observed that women in comparison to men and doctors and nurses in comparison to other emergency employees experienced more anxiety (p < 0.05). | One limitation is that there was no sample size calculation made for the study. Also, the study was conducted in a single institution with small sample size which could impact the generalisability of the results. Some degree of response bias may have been present. | 6/8 |
| Ilczak et al. (2021) Poland [35] | To assess predictors of stress that paramedics, nurses and doctors experience in the face of the COVID-19 pandemic | Online survey | 955 personnel (doctors, nurses, and paramedics) | The predictors of stress in the professional environment included fear of contracting COVID-19, decrease in the level of safety while conducting emergency medical procedures, and marginalisation of treatment for patients not suffering from COVID-19. Additional socio-demographic factors that increased stress among emergency medical personnel were being female and working in the nursing profession. | Use of an online survey may have impacted on response rate and generalisability of findings. There could be other factors predictive of stress aside from what were explored in the study such as the speed and availability of COVID-19 testing and other socio-demographic variables. | 6/8 |
| Jose et al. (2020) India [37] | To assess the burnout and resilience among frontline nurses in the emergency department of a tertiary care center in North India during COVID-19 pandemic | Cross-sectional descriptive design | 120 nurses | The two metrics of burnout, emotional exhaustion and personal inefficacy had a significantly negative correlation with resilience among the frontline nurses in the emergency (r = 0.25, p < 0.05 and r = 0.31, p < 0.01, respectively). A significant negative correlation has been identified between burnout and resilience that informs the role of resilience in alleviating burnout during this pandemic. | The study included nurses working only in the emergency department, while nurses working in other non-COVID areas also may face burnout in varying degrees. Other healthcare providers working in an emergency who are also potential for burnout were not included in our study. Hence, the generalisability of the present study may be limited to only nurses working in the emergency departments. | 7/8 |
| Kelker et al. (2020) USA [20] | To assess the well-being, resilience, burnout, and wellness factors and needs of EM physicians and advanced practice providers (APPs) during the initial phase of the COVID-19 pandemic | Longitudinal, descriptive, prospective cohort survey study | 213 EM physicians and APPs | Concern for personal safety decreased from 85% to 61% (p < 0.001). | The study used an online survey which may subject to susceptible to response biases (i.e., self-selection bias, nonresponse bias, fatigue bias). The response rates ranged from 31% to 53% over four weeks and may impact generalisability of results. Other limitations include: (1) the authors could not assess individual-level change over time due to anonymous data collection; (2) there was lack of race/ethnicity demographic data; and (3) the WBI was used on a weekly basis though the instrument asks questions regarding symptoms “over the last month,” which may lack sensitivity to such degree of change. | 12/12 |
| Impact on basic self-care declined from 66% to 32% (p = 0.009). | ||||||
| Reported strain on relationships and feelings of isolation affected > 50% of respondents initially without significant change (p = 0.05 and p = 0.30 respectively). | ||||||
| Women were nearly twice as likely to report feelings of isolation as men (OR=1.95; 95% CI=1.82–5.88). | ||||||
| Working part-time carried twice the risk of burnout (OR=2.45; 95% CI=1.10–5.47). Baseline resilience was normal to high. Provider well-being improved over the four-weeks (30–14%; p = 0.01), but burnout did not significantly change (30–22%; p = 0.39). | ||||||
| Li et al. (2020) China [25] | To examine the incidence of mental health symptoms and predictors of Post-Traumatic Stress Disorders (PTSD) symptoms among reserve medics working in Wuhan, the capital city of Hubei Province | Empirical, cross-sectional study | 225 reserve medics dispatched to Wuhan | PTSD symptoms and its subscales were significantly associated with age, collegial relationship and mental health status during medics’ service in Wuhan. | Firstly, recall bias may have influenced participants' reported mental health condition while in Wuhan. It is also unclear how long PTSD symptoms may persist or develop in the future. Secondly, study participants were not randomly selected, and findings were not generalisable to the larger population of reserve medics during the pandemic. Thirdly, although the authors surveyed the types of facilities where the healthcare workers were stationed, they did not measure level of exposure, a factor for developing PTSD symptoms. | 8/8 |
| Mental health counselling was significantly associated with PTSD symptoms (OR = 6.30, 95% CI = 2.95–13.46, p 0.01) and having anxiety symptoms (OR = 4.32, 95% CI1.66–11.23, p 0.01) and stress symptoms (OR = 5.95, 95% CI = 1.95–18.15, p 0.01) were associated with PTSD symptoms. | ||||||
| Mental health counselling was a significant factor of avoidance (OR = 4.88, 95% CI = 1.88–12.71, p 0.01) and having anxiety symptoms (OR = 12.67, 95% CI = 4.72–33.98, p 0.01) was significantly associated with avoidance. | ||||||
| Mental health counselling was a significantly factor of intrusion (OR = 5.4, 95% CI = 2.44–11.93, p 0.01) and those with anxiety symptoms (OR = 4.09, 95% CI = 1.69–9.89, p 0.001) were significantly associated with reporting intrusion. | ||||||
| Both depression (OR = 3.4, 95% CI = 1.43–8.1, p = 0.01) and stress symptoms (OR = 3.48, 95% CI = 1.45–8.35, p = 0.01) were significantly associated with hyperarousal. | ||||||
| Munawar and Choudhry (2021) Pakistan [38] | To examine the psychological impact of COVID-19 on emergency HCWs and understand how they dealt with COVID-19 pandemic, their stress coping strategies or protective factors, and challenges while dealing with COVID-19 patients | Qualitative study | 15 frontline emergency HCWs | Findings highlighted a major theme of stress coping, including, limiting media exposure, limited sharing of COVID-19 duty details, religious coping, just another emergency approach, and altruism. A second major theme of challenges included, psychological response and noncompliance of public/denial by religious scholar. | The study was carried out when the pandemic was ongoing and the researchers describe being conscious of not distracting participants from their essential work. Hence, interviews were often paused or interrupted because participants had to attend other emergency calls and duties. Furthermore, the researchers report being conscious of social/physical distancing guidelines, hence, focus group discussions could not be conducted. This study did not report findings of HCWs from private facilities who may have had different experiences of the pandemic and different coping mechanisms. | 10/10 |
| Nie et al. (2020) China [26] | To explore the prevalence and associated factors of psychological distress among nurses working in the frontline during COVID-19 outbreak | Cross-sectional study | 196 ED nurses | Multiple logistic analyses showed that perceptions of having more social support (OR = 0.960, 95% CI = 0.936–0.984) and effective precautionary measures (OR = 0.469, 95% CI = 0.235–0.933) were negatively related to psychological distress. However, working in ED (OR = 3.378, 95% CI = 1.404–8.130), being treated differently (OR = 2.045, 95% CI = 1.072–3.891), concern for family (OR = 2.171, 95% CI = 1.294–3.643), COVID-19-related impact of event (OR = 1.084, 95% CI = 1.052–1.117), and negative coping style (OR = 1.587, 95% CI = 0.712–3.538) were positively correlated to psychological distress. | There were relatively small sample size and low response rates (30–40%). Secondly, the study was based on a self-administered questionnaire so potential for response bias. | 8/8 |
| Data were collected in the early stage of the outbreak so may underestimate prevalence of psychological distress as the impact of the outbreak might be long term, changeable and continuous, on psychological status among nurses at the frontline. Finally, the cross-sectional design limits interpretation of causal relationships between psychological distress and risk factors. | ||||||
| Rodriguez et al. (2020) USA [22] | To assess anxiety and burnout levels, home life changes, and measures to relieve stress of U.S. academic emergency medicine (EM) physicians during the COVID-19 pandemic acceleration phase | Cross-sectional e-mail survey | 426 EM physicians | On a scale of 1–7 (1 = not at all, 4 = somewhat, and 7 = extremely), the median (interquartile range) reported effect of the pandemic on both work and home stress levels was 5 (4–6). Reported levels of emotional exhaustion/burnout increased from a pre-pandemic median (IQR) of 3 (2–4) since the pandemic started a to median of 4 (3–6), with a difference in medians of 1.8 (95% confidence interval = 1.7–1.9). Most physicians (90.8%) reported changing their behaviour toward family and friends, especially by decreasing signs of affection (76.8%). | The primary limitation is the moderate response rate of 57%, which was attributed to using a general e-mail and clinical work overload during the busy early stage of the pandemic and inability to provide gift cards or other incentives. | 7/8 |
| In terms of spectrum effects, the survey was limited to providers at academic institutions and therefore may not reflect experiences of non-academic EM physicians. | ||||||
| Most of the participant sites were in cities in California that had not seen large patient surges as had been seen in other areas of the country at the time of the study. | ||||||
| Song et al. (2020) China [27] | To assess the mental health of emergency department medical staff during the epidemic in China | Cross-sectional design | 14,825 doctors and nurses | Men were more likely to have depressive symptoms and PTSD than women (OR = 1.12, 95% CI = 1.01–1.24). Those who were middle aged, worked for fewer years, had longer daily work time, and had lower levels of social support were at a higher risk of developing depressive symptoms and PTSD. Working in the Hubei province (OR = 1.70, 95% CI = 1.26–2.29) was associated with a higher risk of depressive symptoms, while those working in the Hubei province but residing in another province had a lower risk of depressive symptoms and PTSD. Being a nurse was associated with a higher risk of PTSD. | This study had a cross-sectional design and as the epidemic changed, the mental health of the medical staff may also change. Further research is needed to track the dynamic changes of medical staff’s mental health status. In addition, all participants in the current study are from the emergency department. Due to the diverse working environments and experience of medical staff in other departments, the generalisability of these results to other populations remains to be verified. | 8/8 |
| Vagni et al. (2020) Italy [33] | To identify the direct and mediating effects of hardiness and coping strategies activated by emergency workers on stress and secondary trauma during the COVID-19 pandemic | Online survey questionnaire | 513 Red Cross volunteers | Hardiness and coping strategies, in particular, which stop unpleasant emotions and thoughts and problem-focused, emerged as mediators in reducing the predicted effect of stress on secondary trauma. The mediating effects of hardiness and coping strategies were found to reduce the effect of stress on arousal by 15% and the effect on avoidance by 25%. | This study has several limitations. The first concern is the sampling method, and the second is that the study involved participants from only one region of Northern Italy, Veneto. This region was among the most affected by COVID-19, and this could be a geographical limit, because in other regions of Italy, the pandemic situation was not as serious. The third limitation was the involvement of only a single emergency organisation, the Red Cross. | 7/8 |
| Vagni et al. (2020) Italy [32] | To analyse the relationship of emergency stress and hardiness with burnout among emergency workers | Online survey questionnaire | 494 emergency volunteers | Hardiness showed an effect in reducing emergency stress levels except for inefficacy-decisional stress, emotional exhaustion (β = − 0.277, p < 0.001), and depersonalisation (β = − 0.215, p < 0.001), and simultaneously increased personal accomplishment (β = 0.332, p < 0.001). | One limitation is that the research was conducted through an online survey. Another limitation is that this was the cross-sectional design used in the study, whereas a longitudinal study would allow for a better analysis of phenomena such as the development of symptoms of burnout. The other limitation is the use of a self-reported questionnaire and participants’ lack of knowledge about the presence of previous psychological problems. Finally, the study lacks comparison with other emergency workers involved throughout the national territory and belonging to other organisations, limiting the generalisation of the results. | 7/8 |
| Zakaria et al. (2021) Malaysia [36] | To identify the prevalence of burnout among emergency healthcare worker in this hospital and to identify the factors contributed to the burnout | Cross-sectional study | 216 HCWs | There was weak correlation (r = 0.148) with the years of experience working in emergency department and the level of burnout (p = 0.03). Among the burnout features were fatigue with 52.2% and frequent physical illness and feel unappreciated with 48.6% and 45.9%, respectively. The job-related issues which predisposed to burnout were demand coping with an angry public (70.2%), job overload (63.9%), lack clear guideline or rapid programme changes (54%), and pay too little (53.1%). | There is potential that non-responders might have been suffering from burnout, thus not being willing to participate. Hence, findings might not be an accurate representation. | 7/8 |
HADS = Hospital Anxiety and Depression Scale; PSQI = Pittsburgh Sleep Quality Index.