Assessment of fear, anxiety, obsession and functional impairment of COVID-19 amongst health-care workers and trainees: A cross-sectional study in Nepal [version 1; peer review: 1 approved, 1 approved with reservations]

Background: The emergence of the COVID-19 epidemic threw the world into turmoil. The medical community bore the brunt of the pandemic's toll. Long work hours, and a lack of personal protective equipment (PPE) and social support all had an influence on mental health. Methods: This cross-sectional study was conducted among Lumbini Medical College Teaching Hospital students and employees in Palpa, Nepal. Data entailing their demographic details, pre-existing comorbidities, or death in the family due to COVID-19 was collected using a self-administered survey. In addition, the level of fear, anxiety, obsession, and functional impairment due to COVID-19 was recorded using previously validated respective scales. In total, 403 health-care and trainees participated in our study. The average age of and age (p-value=0.34), these factors were not found to be significantly associated with higher anxiety levels. Nursing students suffered from a significantly greater functional impairment than other medical professionals (mean score=269.15, p-value < 0.001). A moderately positive correlation was observed between fear, anxiety, obsession, and functional impairment scales. This study revealed various socio-demographic characteristics as risk factors for psychological stress in the people related to the healthcare profession of Nepal during the COVID-19 pandemic. A viable answer to this quandary might be adequate psychosocial intervention by health-care authorities, increased social support, and the introduction of better mental health management measures for the front-line medical workers. The authors used the ‘coronavirus anxiety scale’, a tool developed by Lee et al ., 2020. Marital status was not found to have any influence on psychological distress in this study. However, majority of the participants in this study are unmarried. So a relationship with marriage could not be established in this study. This explanation is reasonable. The authors state that death in the family is not contributory to psychological distress in contradiction to other studies, the likely explanation for this discrepancy is the low number of participants with deaths in the family. This is a reasonable explanation. The study design and statistical methodology is appropriate. The participants were anonymized which is important for a study of this nature. Appropriate scales for anxiety were utilized. The study results were concordant with other studies except for marital status and death in the family. The authors explanation for the discrepancy is satisfactory.


Introduction
Viral pandemics and epidemics are notoriously known in history for their public health risks and widescale destruction. From the influenza pandemic in 1918 and its recent outbreak in 2009, 1 severe acute respiratory syndrome coronavirus (SARS-CoV) in 2002-2003 to the Middle East respiratory syndrome coronavirus (MERS-CoV) in 2012 and many more, 2 one thing that has remained constant throughout was the associated high mortality rate and the subsequent social and psychological impact on the general population in the long run. 3 The World Health Organization (WHO) declared COVID-19 as a global emergency in March 2020. 4 Countries around the globe imposed several restrictions, including home confinement, social distancing, following proper hand hygiene, use of face masks, and in severe cases, nationwide lockdowns. 5 These factors combined with the morbidity and mortality associated with the COVID-19 infection have had detrimental effects on the mental wellbeing of the general population but specifically of the front-line medical workers. Previous literature on the SARS-CoV epidemic has implicated medical staff to be particularly susceptible to anxiety, depression, and stress. 6 This may be due to high exposure and, hence, greater risk of contracting the disease or the added workload. These findings can be implicated during the current COVID-19 pandemic because of the same mode of transmission of the infection and a greater patient load than the previous SARS-CoV epidemic.
The advent of the COVID-19 pandemic sent the world into chaos. 5 The medical community suffered the brunt of the pandemic. The lack of medical staff and resources became evident. Long duty shifts, extended work hours, and lack of personal protective equipment (PPE) and social support affected psychological wellbeing. 7 A study in Saudi Arabia conducted during the pandemic reported that 73.5% of the health-care personnel suffered from moderate degree fear and anxiety. 8 Another study conducted by Labrague et al. reported that 37.8% of the nurses suffered from deteriorating mental health. 9 However, the mental health issues experienced by the people related to health-care profession remain the least acknowledged, unaddressed, and untended. 10 Previous literature has focused on estimating the psychological impact of COVID-19 on different sections of the population. 8,11,12 However, the current pandemic has brought to light the necessity to screen individuals who are at high risk for developing mental health issues to optimize their productivity. Thus, in the present study, we aim to assess various psychological distress parameters among the health-care personnel of Nepal and identify personal factors and demographics responsible for predisposing them to a higher risk of developing mental health problems.

Methods
This cross-sectional study was carried out among the staff and students of Lumbini Medical College (LMC) Teaching Hospital, Palpa, Nepal. The sample size computed using OpenEpi 13 was 384, after considering a confidence level of 95% and a frequency outcome factor of 50%. For more robust results, we included 406 participants in the survey, of which 3 participants who did not consent to participate were excluded from the study. The cohort of health-care workers and trainees in the present study included doctors, nurses, other health-care staff, medical students, and nursing students. The present study was approved by the Institutional Review Committee of LMC vide letter IRC-LMC 06-G/020. The questionnaire was in English and disseminated among the medical and nursing students, doctors, nurses, and other health-care staff working at LMC through social media. The first section of the questionnaire was for the consent where it was explained that no financial or material gifts will be provided for completing the questionnaire. The survey did not collect any identifying information of any of the participants and the responses were anonymous. Complete confidentiality of the participants was maintained by not asking them for identifying information like name, working department and post (for employee), year of study (for students) and the email address. Then the participants had an option to choose whether they voluntarily consented to participate or didn't consent. The second section of the questionnaire was accessible only to those participants who had consented. The survey didn't continue for the participants who didn't consent, and the incomplete form were submitted.
The second section of the questionnaire used in the present study consisted of five parts. The first part of this section of the questionnaire was for demographic data and the remaining 4 parts used four different scales which were previously validated. The first part consisted of demographic information such as gender, age, current occupation, and monthly family income. Information regarding respondent's comorbidity, previous contact with any COVID-19 positive case, and if there was a COVID-19 death in their family was also recorded.
The second part of the second section of the questionnaire consisted of the fear of the COVID-19 scale adopted from Ahorsu et al. 14 Fear of the COVID-19 scale is a unidimensional scale with robust psychometric properties and consists of seven items and assessed via five-point Likert scale method (strongly disagree = 1; strongly agree = 5).
The third part of the second section of the questionnaire was used to see the obsession of COVID-19 in the participants. The obsession of the COVID-19 scale (OCS) was adapted from Lee. 15 There were four items to perceive too much coronavirus thought among the participants over the last two weeks. The participants would rate the items using a fivepoint time anchored scale (0 = not at all; 4 = nearly every day over the last two weeks). A score of seven or more signified that the person was overthinking of coronavirus.
The fourth part was the coronavirus anxiety scale. It consisted of five items developed by Lee. 16 The participants would rate the items using a five-point time anchored scale (0 = not at all; 4 =nearly every day over the last two weeks). The participant scoring of nine or more in the questions was considered anxious about the coronavirus. The fifth part was the work and social adjustment scale (WSAS), a measure of functional impairment adapted from Mundt et al., where the participant could score on a scale of 0-8, where 0 meant not at all impaired, and eight meant very severely impaired. 17 A respondent with a total WSAS score above 20 was considered to have moderately severe or severe psychopathology, scores between 10 and 20 were considered to have a significant functional impairment, but less severe clinical symptomatology, and those who scored less than ten were considered to have subclinical impairments.
Data were analyzed using Statistical Package for the Social Sciences (SPSS) version 26 (IBM Corp., Armonk, New York). 18 The Shapiro-Wilk test assessed normality. Descriptive statistics were used to report frequencies and proportions for the categorical responses. The disparity between categorical variables was checked using the Chi-square test. The association between age and vaccine acceptance was assessed through binary logistic regression. In the case of continuous data, Mann-Whitney U and Kruskal-Wallis tests were used. Spearman's rho was used to assess the correlation between the scales, and p-value <0.05 was considered significant in all cases. All the underlying data for the present study is available without restriction. 19 Results

Demographics
A total of 403 health-care workers and trainees took part in our study. The mean age of the study participants was 23AE4 years, and more than half of them were females (n=262, 65%). In terms of the educational level of the participants in the study, nearly half (n=211, 52.4%) of the sample population were medical students. Unmarried individuals constituted a great majority of the sample (n=361, 89.6%). In addition, 166 (41.2%) participants had a monthly family income in the range of 5,000-50,000 Nepalese Rupees. Furthermore, only 13 (3.2%) were suffering from a chronic disease, and just one (0.2%) experienced a COVID-19 related death in the family while a great majority of them (n=376, 93.3%) did not have a positive contact history with a COVID-19 patient as shown in Table 1.

Fear scale
The fear score on the scale ranged from 7 to 35. A higher fear scale score indicated a greater fear towards COVID-19. The mean fear score was 18.7AE5. There was statistically a significant difference between the mean rank scores of males and females (226.02 vs. 157.37), with the males having a higher mean rank score than females. Nursing students had the highest mean scores (mean score= 325.98). A significant association was found between fear score with age (p-value=0.04), gender (p-value <0.01) and occupation (p-value<0.001). The participants with chronic diseases (pvalue = 0.28) and those who had experienced a COVID-19 death in the family (p-value= 0.93) did not show a significant level of fear towards COVID-19. Table 2 summarizes these findings.

Obsession scale
The obsession score on the scale ranged from 1 to 16, with 1 being not obsessed and 16 being highly obsessed with COVID-19. The mean obsession score of the study participants was 2.8AE2.5. Males had a considerably higher mean rank score than females (215.39 vs. 177.13). Nursing students had the highest mean score compared to people of other occupations (289.99). Those with a positive contact history with COVID-19 scored higher than those who did not (243.78 vs. 199.00). A significant association was found between obsession score with gender (p-value=0.001), occupation (pvalue < 0.001), and positive contact history of COVID-19 (p -value= 0.05). It was also observed that age (p-value=0.70), participants suffering from chronic diseases (p-value=0.36), and those who had experienced a COVID-19 death in the family (p-value=0.18) were not found to be significantly obsessed with COVID-19. Table 3 summarizes these findings.

Anxiety scale
The anxiety score on the scale ranged from 1 to 20, with 1 being not anxious and 20 being highly anxious about COVID-19.
The mean anxiety score of the study participants was 0.88AE1.9. Males had a considerably higher mean rank score than females (210.98 vs. 185.32). Nursing students had the highest anxiety score (273.51) compared to other sub-sections of the participants. Health-care workers and trainees suffering from chronic diseases had higher anxiety scores than those without comorbidities (273.38 vs. 199.81). The participants with a positive contact history were more anxious and scored higher than those with no contact history (256.06 vs. 198.12). A significant association was found between anxiety score with gender (p-value=0.009), occupation (p-value <0.001), those suffering from chronic diseases (p-value= 0.008), and those with a contact history (p-value= 0.002). However, factors like age (p-value=0.34) and experience of a COVID-19 death in the family (p-value=0.51) were not significantly associated with higher anxiety levels. Table 4 summarizes these findings.

Functional impairment
The total score ranged from 1-40. A higher score predicted more significant functional impairment. Males had a nominal but significantly greater score than females (212.92 vs. 181.70). Nursing students suffered from a significantly greater functional impairment than other medical professionals (mean score=269.15, p-value<0.001). Factors like age (p-value=0.13), experience of a COVID-19 death in the family (p-value=0.66), contact history (p-value=0.81), other chronic disorders (p-value=0.18) had no significant impact on the functional impairment of the health-care workers and trainees. Table 5 summarizes these findings. Correlation analysis Fear, anxiety, obsession, and functional impairment were positively correlated, with Spearman correlation values (rho) ranging between 0.43 and 0.56; this indicated low to moderately positive but significant relationships (p-value < 0.001) as shown in Table 6.

Discussion
The repercussions the pandemic has on mental health are predictable. Nevertheless, the brunt of the damage endured by people related to the health-care profession is unaccounted for. This study was conducted to evaluate the association of various socio-demographic characteristics of the health-care workers and trainees with various parameters of psychological distress.
In our study we found, older age to be significantly associated with a greater fear of COVID-19. This finding implicates that participants in the higher age bracket were aware of being at higher risk of contracting a severe symptomatic infection which is plausible considering that health deteriorates with the increasing age. There is increased vulnerability of contracting a fatal disease, high risk of hospitalization, and ICU admissions. 20 Our findings concur with the study of Troisi et al., who reported a positive relationship between age and fear level among the health-care personnel of Italy, and the study of Yadav et al. which also reported a greater level of COVID-19 fear among the Nepalese older adults. 21,22 Male gender in our study showed a more significant psychological impact. Male participants in our study were found to have significantly altered levels of all four psychological distress parameters assessed in this study compared to their female counterparts. Our findings were in concordance with findings of studies by Alnazly et al. and Majeed et al., who also reported a greater psychological impact of the pandemic on the male health-care personnel in Jordan and older male adults in Pakistan. 11,23 However, the extant literature reports women to have higher rates of mental health issues which contradicts the findings of our study. 8,24 The differences can be attributed to the study setup, ethnicity, and the cultural norms of the society.
Marital status was not found to influence psychological distress. Alnazly et al. has reported that married individuals have greater levels of fear, stress, and anxiety. Since the majority of the participants in our sample were unmarried, a relationship could not be established. 23 Meraya et al. reported higher family income to be inversely associated with psychological distress. 25 However, no association in our study was established between family income and psychological distress. The differences can be due to the study setup, as people related to the health-care profession were the least liable group of people to face financial issues during the pandemic.
It was also observed that among the sample population, nursing students were found to have the highest levels of all four parameters of psychological distress. Our finding was in line with the findings of Alici et al., who found that nursing students of Turkey were suffering from severe anxiety. 26 In the same context, Huang et al. reported that generalized anxiety and depressive symptoms were more prevalent in the younger than in, the older population. 27 The younger generation fears the pandemic's consequences on their career and has inefficient coping mechanisms. The challenges they face due to distant learning and economic instability contribute to them being more prone to develop psychological distress. 27 We found in our study that a positive contact history rendered the health-care personnel to be more anxious and obsessed, which is plausible because one of the major sources of anxiety among the health-care personnel during the pandemic has been contracting an infection at their workplace and subsequently propagating infection to their families. 28 The healthcare authorities can overcome this concern of people related to the health-care profession by ensuring the availability of PPE and supporting and fulfilling the financial needs of the families of health-care workers and trainees, in case they get infected and have to take time off from work. Our study revealed that participants with preexisting chronic illness were significantly more anxious about the COVID-19 crisis. This is a well-established fact that people with co-morbidities have a higher propensity of contracting an infection and have poorer clinical outcomes. 29 Our finding is coherent with the preexisting literature that also stated the same finding. 30,31 A surprising observation in our study was that, a COVID-19 death in the family was not a contributing factor to psychological distress. This is contrary to the extant literature that reports that a COVID-19 death in the family intensifies psychological distress. 32 The probable cause of this difference may be the inadequate number of participants in our sample reporting a COVID-19 death in the family, resulting in inefficient reporting of the relationship.
In the wake of the pandemic, unpredictability and uncertainty are high. Coupled with the consequences of contracting a severe disease, isolation treatment, and facing the stigma of getting infected, the psychological well-being of the people is bound to suffer. With the health-care workers and trainees in the front line, the stakes for them are even higher. Moreover, a potential for hopelessness, anxiety, and suicide prevails. A possible solution to this dilemma can be appropriate psychosocial intervention by the health-care authorities, enhancing social support, and implementing better mental health management strategies for the people related to the health-care profession.
There were a few limitations in our study. Due to the cross-sectional design of the survey, causal relationships cannot be inferred. Our study is a single-center study, and the generalization of our results is limited. Our sample population was not equally distributed, and most of our participants belonged to middle age and were medical or nursing students, which could have introduced some biases in the results.

Conclusions
This study revealed various socio-demographic characteristics as risk factors for psychological stress in the health-care workers and trainees of Nepal during the COVID-19 pandemic. Enhancing social support and providing a hygienic working environment well-equipped to treat COVID-19 patients and preventing its transmission will prove to be a source of psychological relief for the people related to the health-care profession. Regular psychiatric counseling and an official platform to voice their concerns to the health-care authorities and the government will help mitigate the anxiety and fear of health-care workers and trainees and optimize their productivity.

Open Peer Review
Also, you can situate your findings in the current context of Nepal's health system (federal system) how this pandemic or its outcome may have been influenced by lack of clarity and poor delineation in responsibilities between various tiers. These are again implications to enhance the scope of your findings. Please explore more literature around these themes.

If applicable, is the statistical analysis and its interpretation appropriate? Yes
Are all the source data underlying the results available to ensure full reproducibility? Yes

Are the conclusions drawn adequately supported by the results? Yes
Competing Interests: No competing interests were disclosed. We appreciate the positive comments by the reviewer. Can you add rationales for studying about these constructs and the instruments somewhere in methods? This will clarify the need. And few lines on why quantitative measurement was deemed appropriate, and would qualitative measurement be a limitation? If so, please add that in the discussion as well.
We agree with the reviewer that a qualitative study design with in-depth interview would have provided personal experience of the participants and also helped identify the factors affecting mental health, due to norms of social distancing in the middle of the pandemic, this survey was conducted online with the help of pre-validated questionnaires. We have highlighted this in the discussion section is suggested.

Discussion, Page 17:
"A qualitative study design with in-depth interview would have provided personal experience of the participants and also helped identify the factors affecting mental health. However, due to norms of social distancing in the middle of the pandemic, this survey was conducted online with the help of pre-validated questionnaires." While the study has delved into the stated constructs among health care workers, it would be good to compare and discuss with the studies from Nepal (or elsewhere) who have explored similar constructs among different population (for e.g. in general population) using other methods, to show how that differ and the implications. A bit more specifically, you could compare the constructs between the healthcare workers and the non-health care workers as well. Please explore. Thank you for the suggestion. We have extensively compared findings from our study with other studies that assessed healthcare workers and other populations. The authors feel that comparison of different constructs/methodologies was not within the scope of the current study -and have thus not discussed that aspect

Discussion, Page 15:
"Our findings concur with the study of Troisi et al., who reported a positive relationship between age and fear level among the health-care personnel of Italy, and studies conducted amongst healthcare workers and the general population of Nepal -which also reported a positive association between age and fear of COVID-19. Male gender in our study showed a more significant psychological impact. Male participants in our study were found to have significantly altered levels of all four psychological distress parameters assessed in this study compared to their female counterparts. Our findings wer in concordance with findings of studies by Alnazly et al. and Majeed et al., who also reported a greater psychological impact of the pandemic on the male health-care personnel in Jordan and older male adults in Pakistan. However, the extant literature reports women to have higher rates of mental health issues which contradicts the findings of our study. The differences can be attributed to the study setup, ethnicity, and the cultural norms of the society.
Marital status was not found to influence psychological distress. Alnazly et al. has reported that married individuals have greater levels of fear, stress, and anxiety. Since the majority of the participants in our sample were unmarried, a relationship could not be established. Meraya et al. reported higher family income to be inversely associated with psychological distress. However, no association in our study was established between family income and psychological distress. The differences can be due to the study setup, as people related to the health-care profession were the least liable group of people to face financial issues during the pandemic. In contrary, low socio-economic status was a driving factor for poor mental health among returning migrant laborers in Nepal.
It was also observed that among the sample population, nursing students were found to have the highest levels of all four parameters of psychological distress. Our finding was in line with the findings of Alici et al., who found that nursing students of Turkey were suffering from severe anxiety. In the same context, Huang et al. reported that generalized anxiety and depressive symptoms were more prevalent in the younger than in, the older population. The younger generation fears the pandemic's consequences on their career and has inefficient coping mechanisms. The challenges they face due to distant learning and economic instability contribute to them being more prone to develop psychological distress." You have discussed quite well the findings and the use of scales in various constructs. I recommend you could relate your findings more at proximal level, e.g. how that contributes (affects) the health care workers' preparedness or even health system preparedness? Can we stretch the implications to pandemic preparedness or disaster preparedness for current and future as well?
Also, you can situate your findings in the current context of Nepal's health system (federal system) how this pandemic or its outcome may have been influenced by lack of clarity and poor delineation in responsibilities between various tiers. These are again implications to enhance the scope of your findings. Please explore more literature around these themes. We have positively taken this suggestion and have now discussed the poor preparedness of the Nepalese healthcare system in managing natural disasters and disease outbreaks.

Discussion, page 16:
"Nepal is a lower-middle income country in South Asia with a suboptimal health system preparedness for natural disasters and disease outbreaks. 32 There is a lack of robust surveillance system, diagnostic facilities and management infrastructure. 32 The lack of coordination in the three tiers of governance was evident during the pandemic, specifically lack of healthcare workers, inadequate supply and management of logistics, and diagnostic facilities. 32 The reason for anxiety and depression among the healthcare workers were different from the general public in many aspects. One of the main reasons in the context of Nepal was, the frontline of health-care workers were dutybound amidst the lack of logistics and health safety concerns" Somewhere, can you add a brief explanation about the research site, how is it different to others and what are the social, cultural and even geographical barriers that may (or could) add to the constructs you are measuring? A brief detail to the research site is added.

Methods, page 5:
"This cross-sectional study was carried out among the staff and students of Lumbini Medical College (LMC) Teaching Hospital, Palpa, Nepal, during the COVID-19 pandemic (August 2020). The sample size computed using OpenEpi 13 was 384, after considering a confidence