Introduction

The likelihood of developing a psychological injury versus experiencing psychological growth is influenced by the support received before, during, and after a difficult life event [1]. The COVID-19 pandemic has been an extraordinary challenge for populations around the world. At the focal center of fighting this disease are healthcare workers (HCWs). Protecting them, therefore, must be a crucial component of public health measures aimed at addressing the outbreak [2].

Several reports have discussed the psychological impact of COVID-19 on frontline HCWs [3–5]. The emotional response of HCWs to an outbreak of a disease like COVID-19 is complicated, with potentially long-term mental health implications [2]. Manifestations of psychological distress among HCWs include elevated levels of stress, anxiety, depression [6], severe insomnia, obsessive-compulsive symptoms, somatization [7], post-traumatic stress disorder [8], vicarious traumatization [9], increased risk of developing other mental health problems [1] and in the most vulnerable cases, suicide [10].

Hospitals nationally and globally were struggling with the effects of the pandemic and mental health crisis, including the general population. Women have been found to report more severe symptoms of depression, anxiety and psychological distress than men [3]. Although nurses may be more prone to developing unfavorable mental health outcomes [2], the evidence is mixed, as non-medically trained HCWs may also be at a higher risk for psychological distress during the COVID-19 pandemic in comparison to medically trained personnel [5].

Studies attribute the emotional strain experienced by HCWs to various reasons such as intensified perception of experiencing personal danger, widespread media coverage, inadequate support [2], reluctance to work or contemplating resignation [11], rising numbers of acutely ill patients, anxiety about assuming unfamiliar clinical roles, expanding workloads caring for COVID-19 patients, caring for affected coworkers [3], fear of transmitting the virus to fellow HCWs [8], shortages of medical equipment [3] such as personal protective equipment, limited testing and treatment options for COVID-19, fear of infecting family members due to workplace exposure, the pressure of making emotional and ethical resource-allocation decisions, work-related burnout [12], stigmatization and ostracism for displaying physical symptoms suggestive of COVID-19 infection [13, 14], and limited access to mental health services [3].

It is also important to consider these findings with respect to methodological rigour and quality checks (potentially due to rapid publication of research). For instance, one study reported findings drawn from a relatively small sample size with a low response rate from HCWs, and using measures with single-item ratings [6]. Nevertheless, there is a critical need for healthcare organizations and researchers to prioritize the mental health needs of HCWs serving the community during the pandemic globally [3]. There may even be the risk of the COVID-19 outbreak leading to a ’second pandemic’ of mental health crises in health systems and communities [15]. The implications of this study are needed to develop specialized psychological interventions for HCWs [2, 16], improve relevant organizational and management policies, strengthen and prepare healthcare personnel to provide psychological support and tackle mental health challenges [1, 17], and establish prevention strategies such as screening for psychological distress, as well as providing psychoeducation and targeted support to those at most risk [12].

Methods

Results

Seventy-nine individuals chose the ’disagree’ option at the beginning of the survey, not giving their consent to participate. Out of the 2,019 participants who did provide informed consent, 1,985 HCWs completed the entire survey; these were used to conduct data analyses Fig 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Response rate.

https://doi.org/10.1371/journal.pone.0268976.g001

As shown in Table 1, the sample was predominantly female, aged 30–39 years old, and married. The mean age of the participants was 38.67±9.83 years old (further details on age means included in the supplement, S2 Table). With respect to their work mode in the past 30 days, half of the HCWs had been working within the hospital premises. Nurses and allied health professionals made up most of the sample. Roughly half of the HCWs belonged to clinical hospital departments; out of which (n = 1077), most worked in the Department of Medicine (consisting of subspecialty units for Allergy/Immunology, Endocrinology, Gastroenterology, Infectious Diseases, Internal Medicine, Nephrology, Pulmonary Medicine, and Rheumatology), Critical Care Medicine, Surgery, Pathology & Laboratory medicine, and Oncology (further details included in the supplement, S1 Table) Table 1. In terms of COVID-19 exposure, one-quarter of the HCWs had directly engaged in diagnosis, treatment, or care for COVID-19 patients. However, the majority of them had not contracted the infection themselves, and did not have close loved one(s) who had been affected by COVID-19 Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Sample demographics and COVID-19 exposure.

https://doi.org/10.1371/journal.pone.0268976.t001

The prevalence of mild-moderate psychological distress among HCWs in KSA was high; and severe distress was moderate Table 2. Combined, the incidence of at least mild psychological distress was almost 75%. Derived from the confidence interval, the overall prevalence of at least mild distress was 72.09%. Those who had directly engaged with the care of COVID-19 patients had high prevalence for both mild-moderate and severe psychological distress Table 2. The prevalence of mild psychological distress among those who had no contact with COVID-19 patients was also high Table 2.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Prevalence of psychological distress (N = 1985).

https://doi.org/10.1371/journal.pone.0268976.t002

The proportional odds model for non-modifiable correlates of psychological distress showed that females had increased odds of experiencing one level more of distress than males (OR = 1.23; 95% CI = 1.00–1.50) Table 3. The odds of increasing distress levels for those aged 40–49 and 50–59 decreased in comparison to those aged 20–29, with Ors ranging from 0.49–0.33. There was an 86% decrease in the odds of psychological distress for 60–70 year-olds vs. 20–29 year-olds (OR = 0.14; 95% CI = 0.07–0.27). Participants who reported working as researchers had 46% decrease in odds of elevated distress levels relative to other non-clinical staff (OR = 0.54; 95% CI = 0.33–0.90). Those who had engaged directly with COVID-19 patients were more likely to experience increasing distress than HCWs who did not have contact with infected patients (OR = 1.31; 95% CI = 1.05–1.63). It is clear that there were significantly increased odds of elevated distress levels for HCWs who had someone close to them undergo quarantine, hospitalization and/or death due to COVID-19 vs. those with closed loved ones unaffected by COVID-19. There were also increased odds of elevated distress for HCWs who had to quarantine themselves due to contracting COVID-19 vs. those who had not been infected by the disease.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Proportional odds model with non-modifiable correlates predicting psychological distress (N = 1810).

https://doi.org/10.1371/journal.pone.0268976.t003

After controlling for non-modifiable correlates of psychological distress (reduced model; Table 4), the proportional odds model with modifiable correlates (Model A: financial impact, insomnia, work mode, and other worry-related factors) significantly predicted distress severity. Models with modifiable correlates added individually also showed a significant effect on psychological distress (Models B to R). However, when incremental differences (after controlling for non-modifiable correlates) were considered according to increased significance, the leading correlates of distress severity among HCWs in KSA were insomnia (Model B), worry about COVID-19 generally (Model C), feeling lonely, bored and missing being with friends (Model D), becoming infected by COVID-19 (Model E), the world not returning to what it was before the pandemic (Model F), continuation of the pandemic for a long time (Model G), not being able to practice favourite hobbies due to self-isolation (Model H), infecting close people (Model I), being far from family and loved ones because of the travel ban (Model J), and decline in fitness level and gaining weight due to self-isolation (Model K) Table 4.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Likelihood ratios for proportional odds models with non-modifiable and modifiable correlates predicting psychological distress.

https://doi.org/10.1371/journal.pone.0268976.t004

Discussion

This study was one of the first in the Gulf Council Cooperation (GCC) region to conduct a cross-sectional survey on the prevalence and severity of psychological distress among HCWs in KSA during the COVID-19 pandemic. While several other studies have explored the psychological impact of COVID-19 on HCWs and the general public using different scales in KSA [39–41], our study looked at psychological distress in HCWs at the national level [39, 42], included a larger sample taken from six major hospitals, and examined psychological distress in general rather than focusing on specific disorders [40, 41, 43, 44]. Furthermore, while our results are in line with global findings, Saudi Arabia is culturally different in terms of its psychological landscape [45], and it is possible that religious and cultural beliefs impacted how HCWs responded to the survey. A study conducted in Qatar, a country considered relatively similar to Saudi Arabia in terms of culture and religion, found that Arab survey takers tend to be wary of the reliability and intentions of surveys, whereas for non-Arabs, their behavior towards surveys depends on their willingness as well as the survey’s perceived cognitive and time burden [46]. It has also been found that preference falsification is likely to be present in some Arab countries [47].

We found a high prevalence of mild-moderate and severe psychological distress among HCWs who had engaged in caring for COVID-19 patients, as well as those who were not involved. Gender, age, type of healthcare personnel, contact with COVID-19 patients, having someone close affected by COVID-19 and COVID-19 infection status of the HCW significantly predicted increased levels of psychological distress among the HCWs.

The prevalence of mild-moderate and severe psychological distress was high among HCWs in KSA, similar to the findings of other recent studies [2, 3, 6, 20]. Women were more likely than men to experience increasing psychological distress, consistent with other studies among HCWs [2, 3]. Those in the older age groups (40–70 years old vs. 20–29 years old) had decreased odds of experiencing higher distress. A similar trend was reported in a general population study [48], which posited that younger individuals interacting more with social media potentially encountered more triggers for distress. At the same time, several reports [3, 48, 49] emphasize that the elderly and older individuals still need to be considered with regards to mental health as they are more vulnerable to COVID-19 infection and are witnessing higher mortality rates within their age group.

Several studies conducted in neighboring countries between March and May 2020 also focused on psychological distress levels in HCWs. Authors of a study in Jordan reported depression in 21.2% and anxiety in 11.3% of HCWs; another study in Oman demonstrated moderate-to-severe anxiety in 26% of HCWs. Of note, the difference in figures could be a result of the use of different evaluation tools and the use of different classifications, even if the same scale was used [39].

With respect to type of healthcare personnel, in line with other studies [2, 50], nurses had increased odds (vs. nonclinical HCWs) of experiencing higher distress levels; however, this association was not significant. We found that nonclinical workers are significantly more distressed than researchers, consistent with a previous study [5] that indicated nonclinical workers were experiencing more distress than other types of HCWs during the pandemic. This could be because nonclinical HCWs are more representative of the general population. It has been reported that the distress of nonclinical HCWs is more severe than the distress of frontline workers caring for COVID-19 patients [9]. In terms of COVID-19 exposure, our findings also suggest that HCWs who had directly engaged in the diagnosis, treatment, or care of COVID-19 patients had increased odds of elevated distress levels compared to those who had not been in contact; this aligns with other reports [2, 3]. Given these results, healthcare organizations and hospitals must cater to the mental health needs of HCWs in contact with the infection, and develop specific interventions targeted at both medical and nonmedical staff (interventions will be discussed in further detail below).

Furthermore, those HCWs who had someone close to them in quarantine, hospitalized and/or who had died due to COVID-19, were more likely than those with unaffected loved ones to experience increased levels of distress. Additionally, HCWs who were under quarantine due to a COVID-19 diagnosis (vs. those unaffected) had increased odds of psychological distress, in line with previous studies [51]. These findings were consistent with evidence indicating that sources of distress among HCWs include concerns about health of loved ones, and health of self, wherein the perception of personal danger intensifies [2, 52]. Psychological interventions for HCWs ‒ affected by the pandemic ‒ must therefore ensure addressing risk factors such as potential harm to self and loved ones [17], and the psychological impact related to affected family members [15, 53].

In the multivariate analyses, adjusting for non-modifiable correlates of psychological distress, all modifiable correlates including financial impact, insomnia, work mode, and other worry-related factors remained significantly associated with psychological distress. However, by ranking incremental differences and significance, we found that the correlate most worth modifying among HCWs in KSA was insomnia. Other studies have also emphasized this risk factor and its impact in developing mental health problems including distress, especially during the pandemic [7, 20, 54]. Mental health treatment and intervention approaches would thus be more effective if they pro-actively incorporated plans to tackle sleep quality of HCWs, which can in turn alleviate their distress severity. A study conducted during the COVID-19 pandemic suggests implementing a customized shift system catered to health care workers in high-pressured roles as one possible intervention [55].

Worry about COVID-19 generally, loneliness, boredom and separation from friends, getting infected, infecting close people, and being far from family and loved ones because of the travel ban were other significant correlates that predicted increasing distress levels in the adjusted multivariate analyses. Recent reports affirm that loneliness, boredom, separation from loved ones and uncertainty over disease status can have a dramatic psychological impact during a pandemic [51, 53, 56]. As the transmission of COVID-19 can occur through asymptomatic carriers, psychological distress levels may exacerbate if an individual fears themselves to be the carrier [8]. Concerns about infecting family members (especially those most at risk to contract the infection); [3, 12], and contracting the infection themselves as the number of suspected and confirmed cases continues to increase, contribute to the pressures that the HCWs undergo daily [2, 57]. For these reasons, HCWs, along with the elderly and people suffering from autoimmune diseases, were given priority by the Saudi Ministry of Health when vaccinations and treatments became available [58, 59]. Moreover, with personal travel being curtailed and travel restrictions in place (locally and globally) to limit the spread of infection [58], separation from family results in psychological issues [17, 60]. Du et al. [6] emphasize that strong family support ‒ which due to separation may fall short of being adequate ‒ is vital in increasing frontline HCWs’ resilience to stress during a public health emergency. Various initiatives were launched to support HCWs, including ‘Da’em’, a program launched by the Saudi Commission for Health Specialties to support health practitioners through specialists and various support programs during the pandemic. In addition, King Faisal Specialist Hospital & Research Centre also provided a mental health and psychosocial support hotline in Arabic [53, 61].

Finally, we found that worry about the world not returning to what it was before the pandemic, continuation of the pandemic for a long time, concerns about not engaging with hobbies, and decline in fitness level and weight gain due to self-isolation were some of the other correlates that significantly predicted psychological distress among HCWs in KSA. A similar report from Italy also found that locals experienced high levels of stress due to not having an estimate of how long the pandemic would last and how long their lives would continue to be disrupted [10]. Indeed, collective anxiety and growing fear among communities are the consequence of mass quarantine [62]. In a Chinese study, individuals who exercised regularly reported feeling less satisfied; thus, COVID-19 mental health interventions may need to consider the needs of more physically active individuals, who might experience more distress due to the restrictions posed by the outbreak [21].

Our findings underline the need for psychological intervention plans to address the distress experienced by HCWs in KSA, and are similar to global findings [7, 12, 63]. In addition, our findings can help nationwide strategic decision-makers and policy makers consider the outlined psychological risk factors and prepare for major disasters like the COVID-19 by developing and launching accessible, cost-effective mental health services and recovery programs for HCWs [7, 48]. Telehealth and e-technology services offering virtual peer support, counseling, online resources (e.g., psycho-education), mobile apps, psychological assistance hotlines have been reported to be beneficial in tackling specific sources of distress among HCWs [3, 5, 16, 64, 65]. Previous reports also highlight support provided by healthcare management and direct managers; for example, disseminating clear communications about the disease and logistics related to their work [1, 6, 51, 53] and strengthening personnel via trainings to provide a timely mental health response during emergencies such as the COVID-19 pandemic [17].

This study has some limitations and findings must be interpreted in their light. First, the survey was self-reported and completed by HCWs at a single time point. Longitudinal studies using multiple standardized measures are needed to assess the psychological impact of the pandemic among HCWs. As HCWs may face major changes on a daily basis, monitoring their mental health outcomes over time and prioritizing them is crucial to ensure the quality of care provided to patients [1, 3]. Second, our findings are limited to selected regions of the KSA and cannot be generalized to other differently affected national regions. Third, data about HCWs’ work hours and workload might have provided more detail about the associated distress [6]. Finally, although face validity was carried out to some extent, given the urgency of the topic and shortage of time, most of the questionnaire items did not undergo psychometric testing.

Conclusion

This cross-sectional study found that the prevalence of psychological distress reported by HCWs in KSA was high, ranging in severity from mild-moderate to severe. Younger HCWs, women, those in contact with COVID-19 patients, those with loved ones affected by COVID-19, and those personally affected by COVID-19 were the most at risk of experiencing elevated distress levels. To effectively address psychological distress among HCWs, public health policy makers and mental health professionals must give special attention to risk factors such as insomnia, loneliness, fear of transmission, and separation from loved ones.

Supporting information

Acknowledgments

The SNMHS is carried out in conjunction with the World Health Organization World Mental Health (WMH) Survey Initiative. We thank the staff of the WMH Data Collection Coordination Centre in the Survey Research Center at University of Michigan and the WMH Data Analysis Coordination Centre in the Department of Health Care Policy at Harvard Medical School for assistance with design, instrumentation, fieldwork, and consultation on data analysis of the SNMHS. A complete list of all WMH publications can be found at http://www.hcp.med.harvard.edu/wmh.

We would like to thank all the survey participants, and hospital staff across the country that helped in coordinating this survey’s data collection. We are also grateful to Cielo Mendiola, and Sukina Qanbar at the Biostatistics, Epidemiology and Scientific Computing Department, King Faisal Specialist Hospital and Research Centre, Riyadh for their administrative support.