key: cord-0916290-02d37qqs
authors: Chatterjee, S. S.; Chakrabarty, M.; Dan, U.
title: Mental health of healthcare workers during early phase of COVID19: Variable performance on different factors of stress
date: 2020-09-22
journal: nan
DOI: 10.1101/2020.09.22.20199323
sha: f2e2866e79dc974990fd4e6f9ca386df3b82e12a
doc_id: 916290
cord_uid: 02d37qqs

Background: Risks to healthcare workers (HCWs) escalate during pandemics and they are likely to experience a greater level of stress. This cross-sectional study investigated mental distress among HCWs during the early phase of the coronavirus disease-2019 (COVID-19). Method: 140 HCWs of a tertiary care hospital in India were assessed for perceived stress and insomnia. A factor analysis with principal component method reduced these questions to four components which were categorized as insomnia, Stress-related Anxiety, Stress-related Irritability, and Stress-related Hopelessness. Further statistical analyses were done on these factor scores to identify the predictors and investigate the differences between the different categories of HCWs. Result: Doctors were the most anxious among the HCWs. Both doctors and nurses perceived a greater level of irritability than the other HCWs. Compared to the doctors and nurses, other HCWs were more likely to experience insomnia. Lower age, higher education, female gender, and urban habitat were associated with the perception of anxiety. Older age, quarantine, single marital-status predicted irritability. Female gender, single marital-status, and greater ailments contributed to perceived hopelessness. Quarantine significantly predicted insomnia. Conclusion: Different categories of HCWs might experience disparate mental health problems owing to their heterogeneous socio-demographic backgrounds. Customized and personalized care might prove to be helpful in alleviating their problems.

Coronavirus disease-2019 (COVID- 19) has created an unprecedented situation worldwide and has set forth an array of challenges before us -medical, ethical, and organizational 1 Health care workers (HCWs) are bound by ethics to provide support to patients 2 . HCWs across the world are putting their fullest effort to cope with the pandemic and save lives. However, they are not immune to infection. Consequently, HCWs are equally vulnerable to infection as the rest of the population. In fact, the frontline workers are at a greater risk than the general population. Previous statistics clearly indicate that HCWs make a significant portion of the infected cases 3 .

Owing to increased risk of infection, duty towards patients might tussle with self-preservation and protection of loved ones thereby increasing stress and anxiety of HCWs 4, 5 . Increased duty hours and disrupted biological rhythm during quarantine might lead to insomnia 6 . Inadequate supply of personal protective equipment, problematic media coverage, and stigma might exacerbate stress [7] [8] [9] . In a recent review of six studies, Spoorthy 10 reported that "HCW are encountering a considerable degree of stress, anxiety, depression, insomnia due to the COVID-19 pandemic". Stress and insomnia are not unitary constructs 11, 12 rather these two aspects of mental health are intricately intertwined. Research reports that stress and insomnia exacerbate . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted September 22, 2020. . https://doi.org/10.1101/2020.09. 22.20199323 doi: medRxiv preprint each other and create a vicious cycle impacting long term mental health 13 . Nonetheless, in certain cases, insomnia might be caused by other factors not related to stress. A data-driven factor analysis can help verify if items of stress and insomnia overlap in this particular sample.

India with its several densely populated states, shortage of medical professionals, inadequate equipment, scarcity of health centers, the paucity of testing facilities, sparse surveillance, and poor awareness among masses, failed to contain the disease 14 . Consequently, the pressure on Inadequate number of public health care centers along with the escalating COVID-19 treatment expenses in the private health care centers worsened the situation 16 . The already dwindling patient-doctor relationship 17 reached a worrying level of distrust. Health care workers in general and public health care workers, in particular, suffered acute helplessness.

Stigma, work overload, shortage of equipment, dying patients, distrust, concern for personal safety and safety of the family members pushed them into a mental turmoil.

Recent studies on Indian doctors reported significant mental health problems due to COVID-19 [18] [19] [20] . Most of these studies have acquired data through online surveys that have inherent limitations. Lack of focused groups and selection bias threatens the validity of such studies 21. Here we conducted a pen and paper survey to avoid these limitations 22 .

However, apart from doctors, people working in healthcare facilities such as nurses, ward staff, cleaning staff, porters, and administrative staff are also variably vulnerable 23 and might . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint However, stress is not a unitary construct. It is multifaceted and complex. A global stress score may fail to capture the subtle factors contributing to the elevated stress level. In this study, we conducted a factor analysis on the measures obtained from the perceived stress and the insomnia questionnaires to extract different factors of these two mental health parameters. Subsequently, we investigated if different categories of HCWs have differences in these four different factors of mental health. We also explored if the socio-demographic and the clinicalprofessional parameters could predict these four factors.

The study was approved by the institutional ethics committee (XXXXX/2020/349/10). All participants signed an informed consent form approved by the above committee.

The study was conducted from 20th April to 20th May at Diamond Harbour Government Medical College & Hospital (DHGMC), West Bengal, India. During this time COVID-19 was gradually spreading across India thereby mounting pressure on the health care system. DHGMC was converted into a COVID-19 treatment center, well equipped with an isolation ward, quarantine center, fever clinic, and COVID-19 testing facility.

Of the 550 HCWs employed at DHGMC, 250 randomly selected HCWs, who were on duty during the COVID-19 outbreak, were invited to take part in the study.

Participants having any history of neurological or psychiatric illness were excluded from the . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted September 22, 2020. . https://doi.org/10.1101/2020.09.22.20199323 doi: medRxiv preprint study. After eliminating participants not meeting the inclusion criteria (n=44), non-responders (n=52), and incomplete data (n=14), finally 140 participants were selected for the study.

After factor analysis, factor scores were scanned for outliers. Three outliers were eliminated from further analyses. Tests for statistical significance were done on 137 participants. These 

Demographic information was obtained using a customized demographic data sheet. A questionnaire was designed to assess the participant's level of exposure to COVID patients. Based on the information they were categorized into four groups -severe risk (specimen collection unit, and isolation ward), high risk (chest/medicine outdoor, fever clinic, and emergency), moderate risk (specialist outpatient and inpatient department), and low risk This item was dropped from the final analysis. The final factor analysis yielded four factors.

It may be noted here that the factor structure of a particular tool may vary due to sampling differences 28 . Existing factor analysis data on PSS-10 are based on samples from different cultures and were collected under different socio-economic and health conditions. So, instead of confirmatory factor analysis based on previous studies, a data-driven approach was taken.

'Eigenvalues greater than one' was considered as factor extraction criteria since this is considered to be a reliable technique for factor extraction in an exploratory factor analysis 29 .

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Shapiro-Wilk test for normality was done on the total factor scores and it revealed that the data are not normally distributed. After excluding the three outliers (6, 64, and 125) the data conformed to the normality criteria. Hence rest of the analyses were done on these 137 participants.

A mixed-design ANOVA with groups of HCWs (Doctor, N=55; Nurse, N=45; Ward staff, N=20; and Non-Clinical staff, N=17) as a between-subject variable and the four mental health components (Insomnia, Anxiety, Irritability, and Hopelessness) as within-subject variable was conducted to test for an interaction between the groups and the mental health components. This analysis was followed by independent sample t-tests to determine how the groups differed across the four mental health components.

Stepwise regression was conducted to test if socio-demographic (age, gender, habitat, maritalstatus, education, family, diseases, and media exposure) and clinical-professional variables (duration of service, quarantine, level of risk, contact with confirmed COVID cases, prophylaxis, and use of mask) could predict the mental health components.

The final factor analysis was done on 16 items. KMO of the final model was 0.786 and Bartlett's test was significant ((x 2 (120) = 877.4, p<0.001) confirming that the data were factorable. 30 The diagonals of the anti-image correlation matrix were above 0.5 for all items. Communalities were above 0.5 for all items in the final analysis except P-6 (0.42).

We extracted four factors with eigenvalues above 1. The four components explained 29.6%, 16.0%, 10.0%, and 6.6% of the variance, respectively. The cumulative percentage of variance explained by the five components was 62.2%. The rotated component matrix with the communalities of the items is given in Table 1 .

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Hereafter these four factors will be referred to as Insomnia, Anxiety, Irritability, and Hopelessness, respectively. Factor hopelessness had less than three-item loadings but we retained it as a separate factor because irritability and hopelessness are different aspects of stress. Further analyses were done on these four factor-scores. Independent sample t-tests revealed that compared to the ward staff, doctors were significantly more anxious, (p = .005) irritable (p <.001) and hopeless (p = .001). Nurses were more irritable, (p <.001) and hopeless (p <.001) than the ward staff. Doctors were more irritable than the nonclinical staff (p =.027). Nurses were also more irritable than the non-clinical staff (p =.010).

Nonclinical staff members were more hopeless than the ward staff (p = .008). Ward staff members experienced more insomnia than the nurses (p = .01). There were no significant differences between the doctors and the nurses. See Table 2 for details.

Stepwise linear regression with the socio-demographic variables (age, gender, habitat, marital status, education, family, diseases, and media exposure) as predictors were conducted for all the four factors (Insomnia, anxiety, irritability, and hopelessness). Age (β=, -.431, t = -6.1, p <.001), education (β= .358, t =4.4, p <.001), gender (β = .202, t = 2.7,p = .008), and habitat (β = -.201, t= -2.6,p = .011) predicted anxiety (F(4,132) . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint 

Our study aimed to investigate the different components of perceived stress and insomnia experienced by the HCWs and how different socio-demographic and clinical-professional . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All the 7 questions of the insomnia scale loaded on the first factor. Insomnia was found to be the most important factor and it explained 29.6% of the variance. There was no significant correlation between the factor insomnia and the three factors of perceived stress. This is consistent with Gupta et al. 31 who found no significant difference in perceived stress among three different groups with varying levels of nighttime sleep duration after lockdown due to COVID19. It may be noted that insomnia can be caused by several other factors apart from stress. In this study, quarantine significantly predicted insomnia. More screen time, reduced physical activity, and staying away from home in a quarantine center could contribute to insomnia. Ward staff members were most likely to experience insomnia. Compared to doctors and nurses, other HCWs were more prone to suffer from insomnia.

Stress due to unpredictability has been referred to as anxiety in this study. HCWs with lower age, higher education, female gender, and urban habitat experienced higher levels of anxiety.

In fact, doctors who formed the most educated group among the HCWs were the most anxious of all. As we have seen in several patients, better knowledge and understanding of the disease can engender stress and anxiety 32, 33 . Doctors are perhaps not an exception to this rule. Female HCWs and HCWs with lower age experienced greater anxiety. This is in line with Matud 34 who reported significantly more stress in women even after adjusting for sociodemographic variables. Our result is also in tune with the American Psychological Association's report of . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this this version posted September 22, 2020. . https://doi.org/10.1101/2020.09.22.20199323 doi: medRxiv preprint 2019 35 which states that younger adults and women are the more stressed out. This is also partly consistent with Remes et al 36 who stated that the prevalence of anxiety disorder is higher in women and young adults. However, it may be noted that anxiety referred here is an aspect of stress and we have not used any tool to measure anxiety per se. Nonetheless, these two psychobiological states are reported to have neural as well as behavioral overlaps 37 . Our result is consistent with several other studies that report higher levels of stress in people living in cities compared to rural areas 38, 39 . Fewer years in service and use of prophylaxis was associated with anxiety. HCWs with junior titles were probably less adapted to handle such crises and consequently had higher levels of stress. Higher stress levels could result from the use of prophylaxis 40 . Additionally, people who are more stressed could be more inclined to use prophylaxis.

Stress due to overload has been referred to as irritability. Doctors and nurses scored high on this factor compared to other HCWs. This is consistent with recent studies examining the mental health status of HCWs during COVID-19 7 . Older and single HCWs were more irritable.

This result is quite intuitive. Older people are more likely to succumb to tiredness due to overwork and single HCWs were probably more stressed because they were handling their emotional and physical burden single-handedly. Greater duration of service, quarantine, and a greater level of risk contributed to irritability. This result again is quite expected. Greater duration of service indicates higher age and as already explained older people might capitulate to fatigue and exhaustion more easily than younger people. Moreover, apart from emotional turmoil, quarantine might impose a physical burden as well. Middle-class salaried Indians usually have the privilege of domestic help to take care of household chores. Quarantine could inadvertently repeal this privilege thereby escalating unwonted physical burden and hence stress. This is partly consistent with a study in the general population 41 that was covered by the New Indian Express. HCWs posted in specialties such as a respiratory ward, infectious-. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted September 22, 2020. . https://doi.org/10.1101/2020.09.22.20199323 doi: medRxiv preprint diseases ward, or critical-care ward, where there is a high risk, are plausibly sharing the greatest workload during this pandemic. Consequently, they are probably under greater stress than other HCWs. Wearing the heavy PPE in this hot and humid climate might add to their distress.

Stress due to uncontrollability has been denoted as hopelessness in this study. Female gender, single marital-status, and greater ailments contributed to perceived hopelessness. Ward staff members were found to be the most hopeful among the HCWs. Incidentally, all the ward staff members were males. This is in line with the linear regression result that indicated gender as the most important predictor of perceived hopelessness. Female HCWs were more likely to be perturbed with the feeling of hopelessness. Our result is consistent with studies that report a feeling of powerlessness among HCWs. Females, being more empathetic, are perhaps more likely to feel hopeless when they witness people suffering and dying. As in the case of irritability, unmarried HCWs were found to be more hopeless. This is consistent with a recent study that found lower levels of stress hormones in healthy married adults 42 . Our findings are also in line with Podder et al. 19 who reported higher levels of perceived stress in female and unmarried physicians. HCWs with a greater number of ailments had greater perceived hopelessness. Numerous scientific journals and social media platforms are continuously reporting that patients with lung diseases, diabetes, and heart diseases are at increased risk for severe complications from COVID-19. This awareness and a focus on the uncontrollable could worsen the feeling of hopelessness in HCWs with these ailments. 7 .

In sum, this study revealed the fact that HCWs caring for patients during the pandemic are exhibiting significant mental health symptoms. Several factors are contributing to their psychological distress. Perceived Stress is not a unitary construct. It has three different components and these components have different predictive factors. Female HCWs are more likely to have stress-related anxiety and hopelessness. Doctors and nurses have higher levels . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted September 22, 2020. . https://doi.org/10.1101/2020.09.22.20199323 doi: medRxiv preprint of stress-related anxiety and irritability. Efforts must be made to provide the much needed mental health care to the HCWs working during the pandemic.

This data-driven four-factor solution seems to tap insomnia and the three aspects of stress that Cohen (1994) intended to explore -"….how unpredictable, uncontrollable, and overloaded the respondents find their lives". The study reveals that the HCWs are working with enormous stress and sleep difficulty during the pandemic. Different categories of HCWs are affected differently. Different factors modulate insomnia, stress-related anxiety, stress-related irritability, and stress-related hopelessness.

Small sample size and unequal groups limit the scope of generalizability of the findings of this study. In order to strategically target therapeutic interventions and to establish the possible impact of the pandemic on the mental health of HCWs, confirmation with a larger sample size will be an important next step. The socio-cultural background might influence the intensity and modulate the predictive factors of mental health components. Consequently, while the study from Kashmir 20 reports higher levels of stress among male and married HCWs, we found the female and unmarried HCWs more stressed. Future research needs to tease out how social-cultural factors interact with perceived stress. Although this study is preliminary and exploratory in nature, it generates new hypotheses and addresses an important issue that warrants urgent action.

Mental health issues of HCWs need immediate attention. If this goes on unaddressed it might hamper their performance and subsequent health care delivery. This might not only affect their short-term mental wellbeing but could also have a deleterious long term impact 6, 43 . Well-. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint Key: PS Perceived stress; INS-Insomnia; SUP-Stress-due-to-unpredictability; SOL-Stress-due-to-overload;; SUC-Stress-dueto-uncontrollability . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted September 22, 2020. . https://doi.org/10.1101/2020.09.22.20199323 doi: medRxiv preprint

COVID-19 pandemic: mental health and beyond -The Indian perspective

When health professionals look death in the eye: the mental health of professionals who deal daily with the 2019 coronavirus outbreak

Lives on the line? Ethics and practicalities of duty of care in pandemics and disasters

Severe acute respiratory syndrome (SARS) in Hongkong in 2003: Stress and psychological impact among frontline healthcare workers

Strategic planning and recommendations for healthcare workers during the COVID-19 pandemic

The experiences of health-care providers during the COVID-19 crisis in China: a qualitative study

Factors Associated With Mental Health Outcomes Among Health Care Workers Exposed to Coronavirus Disease

Overview of mental health issues of COVID-19: Need of the hour

Stigma and Aggression Against Health Care Workers in India Amidst COVID-19 Times: Possible Drivers and Mitigation Strategies

Mental health problems faced by healthcare workers due to the COVID-19 pandemic-A review

Symptom Reports in Severe Chronic Insomnia

Stress Concepts

Chronic Insomnia and the Stress System

COVID-19: Challenges and its consequences for rural health care in India. Public Heal Pract

Stigma during the COVID-19 pandemic

Covid doctors under fire from expense-wary patient's kin. The times of India

Changing doctor patient relationship in India: a big concern

Soumitra Das BBB. Attitude, practice, behavior, and mental health impact of COVID-19 on doctors

Comparative analysis of perceived stress in dermatologists and other physicians during national lock-down and COVID-19 pandemic with exploration of possible risk factors: A web-based cross-sectional study from Eastern India

Healthcare Providers on the Frontline: A Quantitative Investigation of the Stress and Recent Onset Psychological Impact of Delivering Health Care Services During COVID-19 in Kashmir

Prevalence and Predictors of Stress, anxiety, and Depression among Healthcare Workers Managing COVID-19 Pandemic in India: A Nationwide Observational Study

COVID-19 online surveys need to follow standards and guidelines: Comment on "Does COVID-19 pandemic affect sexual behaviour? A cross-sectional, cross-national online survey" and "Binge watching behavior during COVID 19 pandemic: A cross-sectional, cross-n

Psychological impact of the COVID-19 pandemic on healthcare workers: a cross-sectional study in China

Perceived stress and its epidemiological and behavioral correlates in an Urban Area of Delhi, India: A community-based crosssectional study

The insomnia severity index: Psychometric indicators to detect insomnia cases and evaluate treatment response

Validity and reliability of insomnia severity index and its correlation with pittsburgh sleep quality index in poor sleepers among Indian university students

Why sample selection matters in exploratory factor analysis: Implications for the 12-item World Health Organization Disability Assessment Schedule 2.0

Multilevel Linear Models

Chapter 2. Basics of linear multilevel models. Multilevel Model

Changes in sleep pattern and sleep quality during COVID-19 lockdown

Disease knowledge level is a noteworthy risk factor of anxiety and depression in patients with chronic obstructive pulmonary disease: A cross-sectional study

Better disease specific patient knowledge is associated with greater anxiety in inflammatory bowel disease. J Crohn's Colitis

Gender differences in stress and coping styles

America 2013: Are Teens Adopting Adults' Stress Habits? PsycTESTS Dataset. 2013

A systematic review of reviews on the prevalence of anxiety disorders in adult populations

Neurobiological links between stress and anxiety. Neurobiol Stress

Cities and mental health

Urbanization and mental health

Safety considerations with chloroquine, hydroxychloroquine and azithromycin in the management of SARS-CoV-2 infection

appropriately to grapple the mental health issues of the HCWs in this difficult time.Personalized treatment for different categories of HCWs might help to maximize efficacy.