Depressive symptoms associated with COVID-19 preventive practice measures, daily activities in home quarantine and suicidal behaviors: Findings from a large-scale online survey in Bangladesh

The present study was a cross-sectional survey conducted between May and June (2020). The study comprised 13,654 participants who resided in Bangladesh. The study’s target population was Bangladeshi citizens who were house-bound in Bangladesh during the COVID-19 pandemic. The inclusion criteria included being (i) aged 18 years or older, (ii) able to read Bangla, and (iii) able to complete the entire survey. The exclusion criteria included being aged under 18 years and not completing the entire survey.

A semi-structured questionnaire was employed to collect data during the survey. An internet-based survey was conducted using a link via Google Forms. A pilot test was performed on 150 individuals to ensure all questions were appropriate and easy to understand. To enable a quick response and to cover all regions in Bangladesh, a voluntary team with more than 200 members were recruited from different parts of the country. This team significantly contributed to collect a large sample by sharing the survey link in their different online platforms (e.g., Facebook, Messenger, WhatsApp, etc.). Initially, 14,353 participants submitted the survey form after obtaining informed consent. Of these, 14,095 participants completed the entire survey voluntarily and anonymously. After eliminating incomplete surveys along with participants below 18 years, 13,654 valid surveys were included in final analysis.

A self-report survey comprised five sections including socio-demographic information, COVID-19 related questions, as well as lifestyle, suicidal, and psychometric measures. Details of these are provided below.

Some basic statistics (i.e., frequencies, percentages, means, standard deviations, chi-square tests, etc.) were investigated. In addition, t-tests or one-way ANOVA tests were performed to determine significant relationships concerning the mean depression scores with all examined variables applying Bonferroni correction (by dividing p-value significance threshold into the number of independent variables [0.05/30 = 0.002]). Finally, COVID-19 related variables, lifestyle and activities in home quarantine, and suicidal behaviors that significantly differed in terms of depression scores were included into hierarchical regression analysis with depression as the dependent variable. The analyses were performed using SPSS software version 25.0.

The study was conducted in accordance with the Institutional Research Ethics guidelines and ethical guidelines involving human participation (i.e., Helsinki Declaration). Formal ethical approval was granted by the Ethical Review Committee, Uttara Adhunik Medical College (Ref. No.: UAMC/ERC/ 13/2020). After providing informed consent, participants completed the survey. The consent form specifically outlined the aims, objectives, nature, and procedure of the study, and clearly stated the right to withdraw data from the survey at any time. All participants were assured that all their data would be anonymous and confidential because no identifying information was recorded (e.g., name, mobile number, address, etc.).

A total of 13,654 participants were included in the final analysis. Of these, 61.4% were male, the mean age was 24.0 years (SD = 5.9), and age ranged from 18 to 65 years, and almost all participants were unmarried (82.4%). The majority had bachelor degree level of education (67.6%), came from nuclear family (78.2%), were higher SES (44.8%), and were from urban areas (62.0%) (Table 1). Two-thirds of participants (67.8%) slept in a normal range (7–9 h) and most (84.0%) did not smoke cigarettes. Additionally, a substantial majority of participants used internet more than six hours daily (31.4%), and 48.9% did not engage in physical exercise during the quarantine (Table 2).

Over two-thirds had a fear that “the coronavirus could infect me” (68.6%). Additionally, two-thirds had a negative impact (decreased comparing pre-COVID-19 situation) on their family income (68.8%), one-fifth of participants had family members who lost their jobs because of the pandemic (18.7%), and a quarter said there was a scarcity of food during the pandemic (24.4%).

Analysis demonstrated a severity depression distribution of 30.5% minimal depression, 26.1% mild depression, 21.4% moderate depression, 14.5% moderately severe depression, and 7.5% severe depression. Based on PHQ-9 cutoff (≥ 10), the estimates of moderate to severe depression were 43.5% with a mean score of 9.5 (SD = 6.7). Almost one in ten participants reported suicidal ideation during the COVID-19 pandemic (8.0%), and 13.7% participants reported past suicidal thoughts prior to the COVID-19 pandemic. Additionally, a minority reported prior suicidal attempts (7.6%), and 3.7% reported a family history of suicide.

With regard to socio-demographic variables, the proportion of moderate to severe depression was significantly higher among (i) females vs. males (47.7% vs. 40.8%, p < 0.001), (ii) participants with comparatively lower vs. higher education (50.8% vs. 41.4%, p < 0.001), and (iii) married vs. unmarried participants (48.0% vs. 42.5%, p < 0 .001) (Table 1).

With regard to COVID-19 related variables, the mean depression score was significantly higher among (i) participants with COVID-19 infected family members, (ii) participants with COVID-19 detection test, (iii) participants with COVID-19 positive result, (iv) participants who had a fear of infection with COVID-19, (v) participants with decreased family income due to pandemic, (vi) participants with family members who lost their jobs during the pandemic, and (vii) participants who had a scarcity of food during the pandemic (Table 2).

The mean depression score was significantly higher among participants who did not engage in good COVID-19 preventive practices vs. those who did, including (i) using handkerchief/tissue during sneezing or coughing, (ii) covering mouth and nose with elbows when coughing or sneezing, (iii) discharging cough/sputum in proper place, (iv) washing hands with soap or other cleaners, (v) using a mask while outside, (vi) discharging tissue in proper place, (vii) maintaining spatial distance, and (vii) maintaining hygiene (Table 2).

With regard to the measures of lifestyle and activities in home quarantine, the mean score was significantly higher among participants who (i) slept more (> 9 h), (ii) were smokers, (iii) used the internet > 6 h daily, (iv) played videogames, (v) engaged in online communication (chat, video calling), (iv) did not read academic books, (vii) did not engage in physical exercise, (viii) did not do household chores, and (ix) did not spend time with their family members (Table 2). With regard to suicidal behavior, the mean depression score was significantly higher among participants with (i) suicidal ideation during the pandemic, (ii) suicidal thoughts prior to the pandemic, (iii) suicidal attempt history and, (iv) a family suicide history (Table 2).

Table 3 summarizes the results of hierarchical regression analysis. Factors that were statistically significant in the group difference analyses (t-tests and ANOVA) were included in a hierarchical regression analysis. COVID-19-related variables were included in Block 1, lifestyle and activities during home quarantine comprised Block 2, and suicidal behavior measures comprised Block 3 (see Table 3). Depression was positively associated with having COVID-19 infected family members, having fear of COVID-19 infection, having decreased family income due to the impact of COVID-19, having family members who lost their jobs during the pandemic, and having a scarcity of food during the pandemic. Furthermore, participants who did not maintain COVID-19 prevention measures were significantly more likely to be depressed than those who did maintain them including discharging cough/sputum in a proper place and washing hands with soaps, sanitizer or other cleaners. Depression was also associated with less or more sleep than normal, being heavy internet users, playing videogames, not studying, having suicidal ideation, and having prior suicidal thoughts. The regression model predicted 16% of the variance in depression (F_(28,5554) = 39.67).

Despite wide variation among community settings, cultural backgrounds, participant perspectives, and survey methodologies, the results reported here broadly concur with previous works, including previous surveys, review papers, and meta-analyses that examined (i) depression, (ii) depression and lifestyle/behaviors, and (iii) suicidal behaviors for general population.

According to the findings of the present study, females were more likely to be depressed than males, and being female has been found as a risk factor for depression in almost all previous studies [6, 42‐44]. However, a study in Pakistan found no significant relationship between sex and depression [45]. The findings relating to levels of moderate to severe depression were significantly higher among those who were married (48.0%) or divorced (47.8%) compared to unmarried participants (42.5%). The condition of being alone after marital separation often affects mental wellbeing and can result in depression [46, 47], and the present study supports and aligns with such findings. Previous literature has also reported that being married (compared to single, widowed or separated) can be associated with depression [48].

Family members of the participants affected by COVID-19 were more highly depressed than non-affected families. Although this is not surprising, no previous study has reported such an association. Obviously seeing loved ones being infected with COVID-19 may act as a traumatic event for an individual which can be a contributory cause of depression [49]. The level of depression was higher among people who had a fear of being infected by COVID-19 which is line in with previous studies [8]. In extreme cases, the fear of COVID-19 has resulted in suicides in Bangladesh and India highlighting the devastating effect on an individual’s mental well-being [12, 50]. This study found depression was positively associated with decreased family income, job loss, and scarcity of food due to the impact of the COVID-19, and several studies also support this finding [51, 52]. Due to countrywide lockdown and shutdown of non-essential industries, millions of people have lost their jobs. This has led to insecurities about livelihood, poverty, starvation, all of which can be contributing to experiencing depression. As noted earlier, several cases of committing suicide had already occurred in Bangladesh and other Asian countries and such extreme actions clearly demonstrate the impact of job losses on the individual’s psychology [53].

This study established an association between depression and COVID-19 preventive practice measures and found higher depression rates among individuals not engaging in COVID-19 preventive practices including hand hygiene practices, face mask use, use of tissue and adequate disposal of it, and other hygiene practices endorsed by WHO to minimize the exposure from highly infectious virus [54].

Lifestyle and activities in home quarantine were significantly associated with depression. Participants who under slept were less or over slept during the COVID-19 pandemic had higher rates of depression in the present study which is consistent with some preceding research that found a correlation between less sleep and depression [53], and an Asian study showing positive association between depression and sleep disturbance [55]. In the present study, excessive use of the internet and playing videogames were associated with depression a finding that has been reported in many reviews and meta-analyses of internet addiction in the area [22, 56], as well as previous studies in Bangladesh [6, 29, 40]. However, during the current pandemic, participants using the internet heavily were also more likely to read various scary, untrue, and/or deceptive news related to COVID-19 that may also contribute to depression [57]. More specifically, depression was higher among participants not engaged or reading academic books, in contrast with other research showing academic pressure correlates with depression [58, 59].

The estimate of suicidal ideation during the pandemic in the present study was 8.2%, which is slightly higher than the 5% prevalence of suicidal ideation found in a pre-COVID-19 study conducted in Bangladesh [33], and 4.3% in Turkey prior to the COVID-19 pandemic [60]. A more wide ranging survey across 17 countries reported 9.2% prevalence of suicidal ideation [61]. Although none of these latter studies were carried out during the COVID-19 pandemic, the finding of the present study demonstrates that the pandemic may possibly be affecting suicidal thinking, as reported elsewhere [62]. A recent study conducted in Bangladesh reported 12.8% suicidal ideation among university students during the COVID-19 pandemic using the comparable methodology [30]. The percentage of individuals with suicidal ideation and past suicidal thoughts were much higher among depressed individuals than non-depressed individuals during the pandemic, and home quarantine and feeling socially isolated may be a contributory issue among people and a possible vulnerability factor for suicide [14].

To the best of the present authors’ knowledge, the study here is a large-scale study in Bangladesh examining depression in relation to different aspects of pandemic issues among the general population including COVID-19 preventive practice measures, and daily activities in home quarantine during the COVID-19 pandemic. Depressive symptoms were associated with COVID-19 preventive practice measures, daily activities in home quarantine, and suicidal behaviors.

This study has some limitations that must be considered when interpreting the results. Firstly, the study was cross-sectional in nature and was unable to determine causality between any of the variables examined. In this respect, a longitudinal study would be more helpful. Secondly, this research adopted an online self-report methodology that may be open to social desirability and memory recall biases. Being carried out online may have increased the veracity of responses compared to offline and/or face-to-face. Although the study was largescale using a convenience sampling technique, it cannot be considered as nationally representative given the higher proportion of males, bachelor degree level of education, and urban residents that participated, and the low mean age of the sample.