Study design and study participants
A cross-sectional study was conducted via web online survey among health workers working in health facilities in Nepal. Data were collected from April 26 to May 12 2020. During the data collection period, Nepal experienced an increase in COVID-19 reported cases from 52 to 217. As of May 12, cases were reported from 19 out of 77 districts of Nepal.
Health professionals working in the management of COVID-19 response, in both public and private health facilities were recruited as study participants through online survey. A network of 25 hub hospitals are designated for COVID-19 management while other hospitals, primary health centres and health posts coordinate with these hub hospitals and run fever clinics for screening of COVID-19 cases. Health professionals included doctors, nurses, pharmacists, diagnostic personnel, paramedics and public health practitioners. A total of 501 responses were received out of which 26 were redundant and hence removed from the analysis. The final sample size of the study was 475.
Data collection methods
The data collection involved two steps: i) identification of survey anchors for participant recruitment online, and ii) survey administration. In the first step, we identified social media platforms and health facility focal persons to recruit the participants followed by non-random sampling of participants interested to participate in the online survey. Online questionnaires on online Google forms were used to collect data from the participants. Study participants were encouraged to fill the online survey form in their leisure. To limit non-health worker’s responses to the online survey, forms were only sent upon invitation to potential participants. The inclusion criteria were health workers aged 18 years and above and living in Nepal, and currently working in COVID-19 management. Participants were excluded if they were below 18 years of age, on leave or unable to participate due to physical or emotional distress.
Study variables
The dependent variables in the study included the status of anxiety, depression and insomnia. The independent variables included information about socio-demographic characteristics and work-related variables. The dependent variables and independent variables are presented in Table 1.
Table 1: Study variables
S.N.
|
Variables
|
Categories of variables
|
Dependent variables
|
1
|
Anxiety
|
Normal (0-7) and Anxiety (more than 7) based on Hospital Anxiety and Depression Scale
|
2
|
Depression
|
Normal (0-7) and Depression (more than 7) based on Hospital Anxiety and Depression Scale
|
3
|
Insomnia
|
No clinically significant insomnia (0-7), Sub threshold insomnia (8-14), Moderate severity (15-21) and Severe clinical insomnia (22-28) based on Insomnia Severity Index. For analysis purpose, a cut-off score of 10 was taken. Absence (0-9) and Presence of insomnia (10 and above)
|
Independent variables
|
|
Socio-demographic characteristics
|
1
|
Age
|
Up to 40 years, Above 40 years
|
2
|
Gender
|
Male, Female
|
3
|
Ethnicity
|
Brahmin/Chhetri, Janajati, Madheshi and others. Adopted from Nepal’s Health Management Information System
|
4
|
Educational qualification
|
Intermediate and below, Bachelor, Masters and above
|
5
|
Profession
|
Doctors, Nurses, Others
|
6
|
Marital status
|
Single, Ever married
|
7
|
Type of family
|
Nuclear, Joint and Extended
|
8
|
Living with child less than 15 years
|
Yes, No
|
9
|
Living with elderly (above 60 years)
|
Yes, No
|
10
|
Having a family member with chronic disease
|
Yes, No
|
11
|
History of medication for mental health problem
|
Yes, No
|
Work related variables
|
1
|
Work role
|
Frontline, second line
|
2
|
Work experience
|
Up to five years, more than 5 years
|
3
|
Type of health facility
|
Primary, secondary and tertiary
|
4
|
Precautionary measures in workplace
|
Sufficient, insufficient
|
5
|
Aware of government incentive for health workers
|
Yes, No
|
6
|
Stigma faced due to COVID-19
|
Yes, No, Do not want to answer
|
7
|
Working in affected district
|
Yes, No (District having at least one confirmed case as affected district)
|
8
|
Working overtime
|
Yes, No
|
9
|
Change in regular job duty
|
Yes, No
|
Data collection measures
Anxiety, depression and insomnia of the participants were assessed using the 14-item Hospital Anxiety and Depression Scale (HADS), and the 7-item Insomnia Severity Index (ISI). Internal consistency of the tool was ascertained by calculating Cronbach's alpha, which was 0.81, 0.72 and 0.90 for anxiety, depression and insomnia respectively and considered sufficient[15, 16].
The HADS is a commonly used tool for measuring anxiety and depression in different settings in many countries including Nepal [17-22]. It has seven items each for measurement of anxiety and depression which are scored from 0-21. The total scores of these tools were interpreted as normal (0-7), borderline abnormal (8-10) and abnormal (11-21). For analysis, score more than 7 was considered as the presence of anxiety and depression. Similarly, the score of ISI which records sleep outcome in the past two weeks was categorised as no clinically significant insomnia (0-7), subthreshold insomnia (8-14), moderate clinical insomnia (15-21) and severe clinical insomnia (22-28) as in studies done elsewhere[8, 23-26]. For further analysis, a cut-off score of 10 was used to categorise the presence or absence of insomnia as suggested by Morin CM et al[27].
Data analysis
Descriptive analysis was done by calculating frequency and percentages for categorical variables and mean and standard deviation for continuous variables. Chi-square test was used to determine the association between categorical independent variables and categorical dependent variables (Additional File 1). To determine potential factors associated with the outcome variable, multivariable logistic regression analysis was performed, adjusted odds ratio (AOR) and 95% confidence interval (CI) were calculated. For adjusted regression analysis, those variables which were significant at a 10% significance level in bivariate analysis were included in the multivariable logistic regression analysis [28]. Similarly, the history of medication for a mental health problem was also fitted into the model regardless of the significance based on the prior knowledge[29]. The Variance Inflation Factor (VIF) was calculated before fitting into the model for each of the psychometric scales which showed no evidence of multicollinearity (less than 1.3).
In the multivariable logistic regression models, the effect of gender, ethnicity, profession, education, living with elderly, family member with chronic disease, precautionary measures in the workplace, faced stigma, worked overtime, awareness about government incentive and history of medication for mental health problem was adjusted to identify the factors associated with anxiety symptoms. Similarly for depression, the effect of age, ethnicity, profession, education, living with children, precautionary measures in the workplace, faced stigma, awareness about government incentive and history of medication for mental health problem was adjusted. Likewise for insomnia,, the effect of age, ethnicity, profession, education, work experience, living in affected district, faced stigma, working overtime, awareness about government incentive and history of medication for mental health problem was adjusted.
Ethics
Ethical approval for the study was given by the Nepal Health Research Council (Reference number: 2192, 315/2020). Written digital consent was taken from study participants prior to completing the survey form. Participants gave their consent by ticking the designated box. Personal identifiers such as name were not collected during the study. The email address collected from the study participants was only used for quality control and not for analysis purposes.