Health care workers depression and associated factors during COVID-19. Health facility-based study in Eastern Ethiopia

Depression of health care workers was related to work absences, resignations, and poor work performance, affecting the quality of patient care and the health care system. It may be at higher risk during the Coronavirus disease (COVID-19) pandemic. There is limited evidence that assesses depression and its associated factors in the study area. Therefore this study aimed to assess depression and associated factors among health care workers (HCW) in eastern Ethiopia. The cross-sectional study design was conducted from October 26th to November 10, 2020. To assess depression among health care workers. Patient Health Questionnaire (PHQ-19) questionnaire was used for the collection of depressive symptoms. The data were analyzed by using STATA version 14 software. Adjusted Odds Ratio (AOR) along with 95% CI was estimated to examine factors associated with depression symptoms using logistic regression analysis. pandemic. health junior health care HCWs perceived susceptibility, perceived severity of COVID-19

Region. The Harari regional state is an enclave inside this zone. The Administrative center of this zone is Harar city. The East Hararghe zone contains four hospitals, and 67 health centers according to the East Hararghe health bureau report [24]. The study was conducted from October 25 to November 15, 2020.

Study design and source population
Health facilities based cross-sectional study design was used. All health care workers from randomly selected health facilities found in the Eastern Hararghe zone were the study population. Health care workers who unable to communicate due to illness were excluded.

Sample size determination
The sample size was calculated using the formula for estimation of a single population proportion (n=[(Zα/2)2 *P(1-P)]/d2 ) with the assumptions of 95% con dence level, marginal error (d) of 0.05, and 50.4% the prevalence of depression among health care workers [25]. Thus, after applying the nite population correction formula and adding 10% of the non-response rate, the nal sample size obtained was 297.
Sampling procedure and sampling technique Two hospitals and eight health centers were selected by simple random sampling method (lottery method) from four hospitals and 39 health centers. The study participants from each selected health facilities were selected by simple random sampling technique by considering proportional to the institutions health professionals' size.

Variables
Dependent variable was depression.
Independent variables were socio-demographic characteristics such as age, marital status, and level of education, work experience, Perceived susceptibility, perceived severity, and perceived bene t.

Data collection and questionnaire
An interview questionnaire was developed to assess socio-demographic characteristics include; age sex, marital status, educational level, type of profession, types of health facility, and worker experiences. The 9-item Patient Health Questionnaire (PHQ-9) tool was used to assess depression symptoms. The questionnaire was validated in the Ethiopian context and it recommended for use [26].
The data was collected by ten nurses. Four MSC health professionals were assigned to supervise the data collection process. Both the data collectors and supervisors were taken 2-day intensive training before the actual work about the aim of the study, procedures, data collection techniques, the art of interviewing, ways of collecting the data, and clari cation. The intensive supervision was done by the principal investigator, co-investigators and supervisors. Completeness, accuracy, and consistency of data were checked at the site of data collection throughout the data collection period. Finally, double data entry was done by two data clerks, and consistency of the entered data was cross-checked by comparing the two separately entered data.

Measurements
The PHQ-9 is the depression module, which scores each of the nine DSM-IV criteria as 0 to 3 for every nine symptoms of depression, "0" not at all, "1" several days "2" more than half the days "3" nearly every day. Scoring was done by counting the number of boxes checked in a column. Multiply that number by the value indicated above, then add the subtotal to produce a total score. PHQ-9 total score for the nine items ranges from 0 to 27. Scores of zero, 1-4, 5-9, 10-14, 15-19, and 20-27 represent cut-points for no depression symptoms, minimal, mild, moderate, moderately severe, and severe depression, respectively [27]. For logistic regression analysis we have categorized those who had no symptom as have no depression and coded as "0" and the rest who had at least one symptom categorized as had depression symptoms and coded as" 1" The age of the mother was recorded based on maternal response later and was grouped as 19-24, 25-29, 30-34, 35-39, and ≥ 40 with codes 1,2,3,4,and 5, respectively, for analysis. Work experience was grouped as < 2 years, 3-5 years, 6-10 years, and >10 years with codes 1, 2, 3, and 4, respectively, for analysis. Data for perceived susceptible was collected by interview using a single question "How likely do you think it is that you will develop COVID-19 during your providing care?" liker scale of ve were used and labeled as "Strong unlikely", "unlikely", "Neutral", "likely", and "Strongly likely" and coded with 1, 2, 3, 4, and 5 respectively. For logistic regression analysis, we categorize as strong unlikely, unlikely, and neutral as no perceived susceptible, no susceptible coded as "1" and susceptible coded as "2". With a similar pattern. Data for perceived severity was collected by interview using question with a single question "Getting COVID-19 in the future worries me and It is important for me to prevent getting". Likert scale of ve was used and labeled as "Strong unlikely", "unlikely", "Neutral", "likely", and "Strongly likely" and coded with 1, 2, 3, 4, and 5 respectively. For logistic regression analysis, we categorize as strong unlikely, unlikely, and neutral as no perceived susceptible, no susceptible coded as "1" and susceptible coded as "2".
Interpretation PHQ-9 questionnaire is not a screening tool for depression but it is used to monitor the severity of depression and response to treatment. However, it can be used to make a tentative diagnosis of depression in at-risk populations (Table 1). Double data entry was made using the Epidata 3.1 software. Then after validation was done the data exported to the STATA statistical package version 14 for further analysis. Descriptive statistics were used to summarize the variable. Continuous variables like age and work experience were rst transformed into categorical variables before analyzed. Initially, the crude odds ratio (COR) along with a 95% con dence interval was estimated to assess the association between each independent variable and the outcome variable. Multicollinearity was tested using the Variance In ation Factor (VIF) test and the tolerance test.
No multicollinearity problem was found. The Hosmer-Lemeshow goodness-of-t tests were used to test for model tness [25]. The logistic regression model was used to assess the association between predictor variables and the outcome variable depression symptoms. Adjusted Odds Ratio (AOR) along with a 95% con dence interval was estimated to assess the strength of the association. Statistical signi cance was declared at a p-value ≤of 0.05

Ethical consideration
The study was approved by the Haramaya University, College of Health and Medical Sciences Institutional Health Research Ethics Review Committee (IHRERC). The permission and agreement consent was obtained from East Hararghe Zone, participating District health bureau, and health care facilities o ce prior to the study. The study participant was informed about the purpose of the study, their right to refuse, and written and signed voluntary consent was obtained prior to data collection. Based on the interpretation and recommendation HCWs who had depression symptoms were obtained management.

Results
Socio-demographic characteristics of the respondents A total of 265 health care workers were participated in this study. The mean (± SD) age of HCWs was 29.29 (± 6.40) years. The mean (± SD) HCWs work experience was 6.25 (± 4.98) years. Of these, 69(26.04%) were working in outpatient department, 49 (18.49%) were working in mother and neonatal health, the majorities 118 (44.53%) were nurse in profession, and more than two-third 183(69.06%) were married (Table 2).     (Table 6).

Multicollinearity test
Multicollinearity was tested using the Variance In ation Factor (VIF) and tolerance test. Variance In ation Factor for all variables was less than ve. Variance In ation Factor below 5 is believed non-signi cant multicollinearity and therefore there is no need for correction [28] (Table 7).

Discussion
In this study, overall, 276 (66.42%) of the study participants have had at least one self-reported depression symptoms. Factors signi cantly associated with depression symptoms were being a female, work experiences, perceived susceptibility, and perceived severity.
The depression symptoms reported in this study was consistent with those reported in studies conducted elsewhere [29][30][31][32][33]. This may be due to fears and concerns for self, family members, problems with children and spouses, and di culties entailed in working at home are some of the stressors during the Covid-19 pandemic [34]. This may also due to, Ethiopia is one of low income countries which left HCWs with high fear of COVID-19 due to lack of personal protective equipment(PPE) and shortage of human power to handle ever increasing the case [14,35], and low preparedness to compact the pandemic [35 -37].
As some bodies experience stress, adrenal glands make and release cortisol hormone into the bloodstream. Prolonged elevated cortisol level leads to the manifestation of depression [38][39][40] and minimize the body immunity [41]. Health care providers at higher risk of experiencing severe disease and even may fatal. ICN con rms 1,500 nurses have died from COVID-19 in 44 countries and estimates that healthcare worker COVID-19 fatalities worldwide could be more than 20,000 as of October 28, 2020 [42]. Without addressing mental health services, we risk not having a workforce left to care for us.
The study showed depression symptoms were higher among female health care workers. This is supported by studies conducted in China [19,43], Gulian [44], Egypt [45] and Ethiopia [46]. Women are among the highly vulnerable population groups during a pandemic due to disrupting the food system and other economic activities. A study in India showed COVID-19 case fatality rate among men were 2·9% and 3·3% among women [47]. Study showed that, there was gender difference in stress and coping styles, numerous studies have determined that women nd themselves in stressful circumstances more often than men, and their coping style is more emotion-focused than that of men [48]. Social roles also seem relevant in the stressful life experiences of women, especially in low-income countries. There can be sex differences in the use of coping strategies which is important for the mitigation of depressive symptoms [49]. The study shows that certain psychological and social determinants were associated with increased depressive symptoms, women are disadvantageous in low-income country, COVID-19 enhances the vulnerability to the mental health challenge [50]. Measures must be put into place to ensure the protection of women on the frontlines while reducing COVID-19 deaths and adverse health effects among displaced populations.
This study also showed work experience was signi cantly associated with had symptoms of depression. Health workers who had less work experience had developed more depressive symptoms. This found to be in line with a nding study done in Nepal [51]. The study showed a lack of communication skills with patient, among health professionals and colleague, medical knowledge, and skills associated with depression symptoms [52], less experience may weaken self-regulation ability and copying mechanism for stressor [22]. Training on knowledge and skill and schedule for activities mixing senior professionals who had skill and experience with junior health workers may bene t.
Perceived susceptibility (perception of the risk of acquiring COVID-19) was associated with participants who had symptoms of depression. This is consistent with the study nding in China [53,54]. Also, participants who perceived the severity of COVID-19 were six times more likely to develop depression compared with their counterparts. This is in line with a study done in China [19]. Perceived susceptibility may be related to shortages of personal protective equipment (PPE) or other essential equipment, low community role in practice to protect themselves and the play their role in mitigating the transmission including to the health care worker [55]. Contradicting of belief in susceptibility and severity, with the low accessibility of protective measure, less concern of the community and the health manager may put health care provider at risk of experiencing depressive symptoms. Health care may also face moral dilemmas in decisionmaking around the provision of care with limited resources [6].
Depression symptoms were related to work absences and reduced work performance. The study investigated that workers with depression may bene t from care involving medical and psychological interventions [22]. World Health Organization identi ed protecting the mental wellbeing of healthcare workers caring for people with COVID-19 has been recognized as crucial for the long-term capacity of the health workforce [23]. However, most health care workers experiencing depression were not planning to seek help and not aware of workplace help plan [22]. Prioritizing the wellbeing of health care workers must be at the forefront of the action plan to tackle the coming wave of this pandemic. We suggested that comprehensive psychological assistance should be provided to support the mental well-being of healthcare workers.

Strengths, limitations, and future directions
The study conducted in multiple health care facilities including Hospitals and Health Centers found in town and rural areas. However, the cross-sectional study is important to be aware of the predictive, there is generally no evidence of a temporal relationship between exposure and outcome because the exposure and outcome are simultaneously assessed. The study includes only depressive symptoms, other mental health problems were not included. This study also not include staff who had contact with COVID-19 patients. We suggest a study that includes health professionals who had contact with COVID-19 patients, assessment of other mental health problems, longitudinal studies that assess the prolonged implication on HCWs need further investigation.

Conclusion
A high proportion of health care providers experienced depression symptoms. The mental health of healthcare workers should be protected during the COVID-19 pandemic. Female health workers, junior health workers, have less work experience and perceived susceptibility, and perceived severity of COVID-19 need attention. Declarations