1 Introduction

The SARS-CoV-2 virus, similar to previous health catastrophes like the Spanish influenza (H1N1), became a global concern that endangered the overall wellness of women and girls worldwide. Even though men reported more severe COVID-19-related health cases, women were at a higher risk for long-term health consequences related to their maternal and reproductive health [1,2,3,4]. The pandemic has exposed pre-existing gaps in achieving sustainable development goals, including poverty eradication, health, and well-being, particularly within socioeconomically deprived communities [5, 6]. In Bangladesh, slum dwellers in cities like Khulna, with a population of 200,000 spread across 520 slums, have faced long-standing structural inequalities during the COVID-19 pandemic [7].

At the beginning of March 2020, the Institute of Epidemiology, Disease Control, and Research in Bangladesh announced the confirmation of the first three cases of COVID-19. From March to May 2020, the country reported a total of 182 COVID-19 deaths [8]. Subsequently, in June 2021, the government declared lockdown after the health directorate reported the second-highest daily COVID-19 death count of 108, the most since the virus’s inception in March 2020. For the first time, the government took advantage of the phrase “lockdown” during nationwide COVID-19 measures to emphasize the severity of the upcoming restrictions. The restrictionsallowed only ambulances and other medical service vehicles on the roadways and prohibited all vehicles carrying goods other than necessities [9]. Since the pandemic began, Bangladesh has reported 29,431 deaths and 2,036,527 COVID-19 cases as of November 2022 [10]. The global COVID-19 crisis is not merely a trial for global healthcare systems but also exposes the darker aspects of human nature. As the pandemic restricted movement and intensified social and economic pressures, gender-based violence experienced an alarming surge [11].

The COVID-19 pandemic has resulted in direct and indirect health effects on women and men [12]. Although the rate of COVID-19 cases was higher among men than women, later studies reported that the gender differences in COVID-19 cases were possibly due to underreporting in many countries where women have fewer rights and access to resources than men [13]. The COVID-19 outbreak resembles the influenza pandemic of 1918–19, in which the virus most severely affected disadvantaged communities and caused higher rates of deaths in denser urban areas [14, 15]. There is scant data on the effects of the Spanish influenza, which occurred during World War I, and documentation of its psychological and societal consequences was rare in the early twentieth century. Children and adolescents who lost their parents due to the Spanish Flu struggled with various mental health issues, exacerbated by wartime poverty [16, 17]. Similar to the Spanish Flu, poverty emerged as a vital issue during COVID-19, with individuals experiencing high stress levels due to financial hardship and self-quarantine measures [18]. In countries like Bangladesh, the pandemic has affected the livelihoods of both men and women living in slums, where many rely on informal and low-paying jobs that have been severely disrupted due to movement and economic activity restrictions [19]. The economic consequences of COVID-19 exacerbated the crisis among the poor Bangladeshi people, with per capita income dropping by 82% from $1.30 to $0.32 (US) for individuals living in slums [20, 21]. Due to economic challenges, Bangladesh has seen an increase in COVID-19-related suicide cases. However, the exact number of suicide cases was uncertain, as families in Bangladesh are reluctant to disclose their loved ones' deaths as suicide news to avoid the social and legal complications associated with such deaths [21]. During the COVID-19 pandemic, research on impoverished residents of Dhaka found that depression symptoms were more prevalent among females, and reduced household incomes were associated with factors such as lack of primary education, unemployment, food scarcity, and depression [22]. The pre-existing co-morbidities, insufficient water sanitation and health facilities, insecure livelihoods, unhealthy environments, and lack of awareness have increased the vulnerability and transmission risks of slum dwellers [23, 24]. In addition, the insufficient healthcare system and manpower, lack of testing kits, and overall lack of comprehensive management policy during the pandemic have aggravated the infection rate in urban slum areas [23, 25]. In 2021, Akter et al. discussed the vulnerabilities and adaption practices of the pandemic in the slum areas of Khulna City, Bangladesh [26]. Four years after the transition from MDGs to SDGs, the majority of the services in the slum of Rupsha were inadequate and failed to satisfy residents, with the water service provision being particularly deficient [7]. Moreover, climate migrants residing in urban slums faced significant challenges related to water scarcity, inadequate wash facilities, and livelihood opportunities [27]. Most research has indicated that different communities face distinct forms of inequality, suggesting the need for local strategies to enhance community resilience during pandemics [14]

Natural disasters wreak havoc on over one-third of Bangladesh's coastal communities yearly; notably, the Khulna division in southern Bangladesh is a hotspot for such catastrophes [28]. Due to the frequent climate-related disasters in this coastal area, many people migrate to Khulna City and settle in informal settlements. During pandemics, men and women in these vulnerable communities may struggle differently regarding their duties and responsibilities. In a patriarchal society, women and men experience any disaster event (man-made or natural) differently according to their gender perspective [29]. Understanding the gender-specific consequences of the COVID-19 pandemic among slum dwellers is crucial to developing fair and effective policies for future pandemics. The focus of this research is to investigate the socioeconomic and gender-specific mental health impacts of COVID-19.

2 Materials and methods

2.1 Study design and sampling method

The study was conducted among individuals living in the slums of Khulna City, Bangladesh, to assess the impact of the pandemic on their socioeconomic and mental health status concerning gender. The population size of the city is around one and a half million, and approximately fourteen percent of the inhabitants live in 520 slums [7]. A cross-sectional study was carried out with 404 individuals (> 16 years) living across 8 wards within the city. A structured survey questionnaire was developed in English using Kobo Toolbox, then translated into Bangla with a forward–backward translation method for urban slum residents, and pretested before being finalized for use in interviews. Subsequently, a pilot survey was carried out to evaluate the acceptance of the questionnaire and transparency among the target group. The questionnaire underwent minor revisions following the pilot test and the pilot survey responses were omitted from the final analysis. The questionnaire is structured into four sections: (i) Socio-demographic characteristics (e.g., gender, age, family size, education level, employment status, monthly income, source of drinking water, toilet facilities, etc.), (ii) Health condition during and after COVID-19 pandemic, (iii) Socioeconomic impacts during the pandemic, (iv) Mental health status during the pandemic.

The study utilized a statistical formula, as proposed by Bartlett et al. [30], to derive the minimum sample size required to obtain a representative sample of households in each informal settlement. We estimated the sample size for the population means by utilizing the following equation:

\(n_{0} = \frac{{Z_{{{\raise0.7ex\hbox{$\alpha $} \!\mathord{\left/ {\vphantom {\alpha 2}}\right.\kern-0pt} \!\lower0.7ex\hbox{$2$}}}}^{2} P\left( {1 - P} \right)}}{{d^{2} }}\) where, \({n}_{0}\) is the initial sample size, \(Z_{{{\raise0.7ex\hbox{$\alpha $} \!\mathord{\left/ {\vphantom {\alpha 2}}\right.\kern-0pt} \!\lower0.7ex\hbox{$2$}}}} = Z_{{{\raise0.7ex\hbox{${0.05}$} \!\mathord{\left/ {\vphantom {{0.05} 2}}\right.\kern-0pt} \!\lower0.7ex\hbox{$2$}}}} = Z_{0.025} = 1.96\) standard normal value. The study used a 95% confidence level (i.e., \(\alpha\) = 0.05), an acceptable error of margin, d = 0.05, and \(P=\) the percentage of people living in urban slums who were affected mentally and socioeconomically during the Covid-19 pandemic, approximately 48% and 50% respectively, as determined from pilot survey. This procedure resulted in a minimum sample size of 384. Finally, considering 5% non-response thetotal of \((384/\left(1-0.05\right))\approx 404\) households was estimated. All primary data was obtained within a predetermined timeframe from July to August 2023, and to ensure the authenticity and validation of the data, a random verification process was conducted by re-interviewing 5% of the respondents. This study was conducted across eight wards of Khulna City Corporation (KCC), and a two-stage sampling method was utilized to obtain a sample of 404 households in the slum area, ensuring proportional allocation of selection based on their relative size within the slum area [31, 32]. First, using Cochran's formula, we estimated that 404 participants were needed to achieve a 5% level of precision. Subsequently, using probability proportional to size (PPS) sampling, we obtained the required sample households for each slum. The study data were collected from the selected study areas (15 slums), where the sample size for each selected slum was determined based on its proportion to the total slum population (Supplementary file 1).

2.2 Data collection tools and study measure

Eight graduate-level students from Khulna University were recruited as data collectors and conducted face-to-face household surveys using the survey tool ‘Kobo Toolbox’. Socio-demographic variables considered in this study were: gender (male, female), age (≤ 25, 26–40, > 40), marital status (married, unmarried), years lived in the slum (≤ 15, 16–30, > 30), monthly household income (BDT) (≤ 7 K, 7001-12 K, > 12 K), Education level (no formal education, primary, secondary or higher), family size (≤ 3, 4–5, > 5), number of school going children in the family (zero, one, two or more), have any family member who is suffering from chronic illness ( yes, no). Kuppuswamy’s Socioeconomic Scale (SES) was used to evaluate the socioeconomic status of slum dwellers based on the information of the head of the household. The participant was first asked if he or she was the head of the household; if the response was negative, then the participant inquired about the household head's details, including the relationship with the participant, household income, employment status, and education level of the household head [33]. From that scale, the socioeconomic status was categorized as ‘extreme poor or below the poverty line’, ‘poor’, and ‘lower middle’ group. The hygiene and surrounding environment-related variables considered in this study were: sanitation facility (unimproved, community latrine, latrine shared between two or more households, family latrine), garbage disposal (fixed place or Khulna city corporation dustbin, no fixed place), Source of drinking water (unimproved/public standpipe, tube well). Furthermore, the pandemic-related variables considered in this study were: experienced food scarcity during COVID-19 (yes, no), received support from NGOs or Govt. during the pandemic (yes, no), taken additional responsibilities due to COVID-19 (yes, no) [22, 26]. The outcome variable ‘socioeconomic impact of the pandemic’ was evaluated by inquiring., five Likert scale questions (‘Household income had decreased since the outbreak of Covid-19’, ‘Trouble accessing necessities like food and medicine because of the pandemic’, ‘Unable to seek medical care due to financial difficulties during the pandemic’, ‘Borrow money to meet daily needs since the outbreak of the pandemic’, ‘Pandemic affected my ability to pay rent or mortgage’) as previously reported [22]. Likewise, four Likert scale questions (‘I had felt down, depressed, or hopeless since the outbreak of Covid-19’, ‘I had difficulty falling asleep(< 30 min) during the outbreak of Covid-19’, ‘Because of unpleasant situations, I felt anxious or helpless during the pandemic’, ‘I felt stressed managing livelihood during the pandemic’) was asked to access the mental challenges faced by slum dwellers during the pandemic [34, 35]. The rating was given on a 5-point scale from 1 (Never) to 5 (Always). For a more accurate measure, we combined the responses as a mean score to form a single variable, which was interpreted as; 1 to 1.8 = Never, 1.81 to 2.60 = Rarely, 2.61 to 3.40 = Sometimes, 3.41 to 4.20 = Often, 4.21 to 5.00 = Always [36, 37]. For analysis purpose the responses were categorized into three categories as; ‘Mild’, ‘Moderate’ and ‘Severe’. For the outcome variable ‘impact on mental health during COVID-19’, the mean scores lower than 25% of the distribution were categorized as ‘Mild’; scores at or between 25 and 75% were categorized as ‘Moderate’; and scores above 75% were categorized as ‘Severe’.The internal consistency and reliability of item scores were evaluated using Cronbach's alpha. The scores range from 0 to 1, with higher values representing greater consistency [38]. In this study, the Cronbach’s alpha values for the Likert scale questions on the Socioeconomic Impact and Mental Health impact during the pandemic were 0.828 and 0.856, respectively.

2.3 Data entry and statistical analysis

After the completion of data collection, the data was retrieved from Kobo Toolbox in the format of an SPSS file, cleaned, and imported into STATA Windows version 17.0 (StataCorp. LP, College Station, TX, USA) for further data cleaning and analysis purposes. Socioeconomic impact and mental health conditions are two response variables in this study. The relationship between the selected response and explanatory variables is explained by using the Logistic regression model [39]. To measure the variables that are linked with the response variables, we utilized the chi-square and the gamma test [40]. Those variables with p < 0.25 were further selected for the multivariate model. We checked for multicollinearity before performing logistic regression. No variable had a VIF (Variance Inflation Factor) score greater than 2.5 or a tolerance value lower than 0.1, indicating the absence of multicollinearity in our model. For multivariate analysis with ordinal logistic regression models, there is a crucial assumption regarding ordinal odds; that is, parameters should not alter for various categories, in other words, parallel slopes. For statistical models that fail to satisfy the proportional odds assumptions, the generalized ordered logistic regression with the auto fit (also known as partial proportional odds (PPO) model is a more suitable option as it covers the gap between ordered and non-ordered modeling frameworks and is more parsimonious than multinomial logistic models [41, 42]. The research study utilized the AUTOFIT option with GOLOGIT2 to generate a partial proportional odds model, in which the parallel-lines constraint is relaxed for variables that violate the assumption, and the parallel-lines constraint is maintained for the variables that uphold the assumption [43, 44]. In other words, the same coefficients are seen across models if they do not violate the parallel assumption. Data were analyzed using STATA software, and with a 95% confidence interval, the significance level was set at p < 0.05.

3 Results

The findings of the study are presented in this section where a total of 43.6% of men and 56.4% of women participated. Occupational situations of the city slum’s women and men during COVID-19 are presented in Supplementary file 2. It is evident from the figure that female slum dwellers experienced higher rates of mental health severity during the pandemic than males. Regarding slum dwellers socioeconomic impact due to the pandemic, the collected sample reported a higher percentage of males experienced job loss or reduced working hours compared to females. The bivariate analysis in Table 1 demonstrates a significant correlation between ‘family member suffering from chronic illness’, ‘employment status’, ‘education level’, ‘smoking status’, ‘experienced food scarcity due to COVID-19’, and ‘taken additional responsibilities due to the pandemic’ with the socioeconomic impact status.

Table 1 Association between selected variables and COVID-19 impact on socio-economy using Chi-square and gamma test
Full size table

Notably, the severe socioeconomic impact due to COVID-19 was highest for unskilled workers/ labor/ rickshaw pullers. Among the slum dwellers, the severe socioeconomic impact due to COVID-19 was 17% higher for male participants and 16% higher for those who smoke tobacco. Table 2 demonstrates a significant correlation between ‘family member suffering from chronic illness’, ‘Received support from NGO or Govt. during the pandemic’, ‘Experienced food scarcity due to COVID-19’, ‘Difficulty accessing basic needs during the pandemic’, ‘Unable to seek healthcare due to financial difficulties during the pandemic’, and ‘Borrow money to meet daily needs during the pandemic’ with the mental health severity during COVID-19 for both men and women.

Table 2 Association between selected variables and COVID-19 impact on mental health stratification of gender in the urban slum using Chi-square and gamma test
Full size table

Furthermore, mental health was associated with the ‘employment situation during the pandemic’ for male slum dwellers, while no significant association was observed for females. The proportion of severe and moderate impact on mental health was found to be higher among females with chronic illness (32.1% and 64.3%), having no fixed place for garbage disposal in the slum area (31.3% and 58.9%), used community latrine during pandemic (29.7% and 57.8%), and taken additional responsibilities (e.g. financial and family responsibilities) due to the consequence of the pandemic (34.7% and 57.1%). For male slum dwellers the severe and moderate impact on mental health was found higher for those who experienced food scarcity due to COVID-19 (17.6% and 66.2%), lost job or working hours reduced during the pandemic (14.3% and 71.4%), not received support from NGO or Govt. during COVID (14.9% and 52.1%), and borrowed money to meet daily needs during the pandemic (13.1% and 61.5%).

The partial proportional odds model (PPOM) in Table 3 revealed that certain variables, such as gender, family member suffering from chronic illness, experienced food scarcity, the 7001-12 K monthly household income, as well as slum living year of the category greater than 30 and employment status of the category self-employed, violated the assumption of parallel lines.

Table 3 Partial proportional ordered logistic regression models on the determinants of socioeconomic impact due to COVID-19
Full size table

Consequently, the odds ratios for these variables were permitted to vary (AOR1 ≠ AOR2) across models. Based on the results of the PPOM, several factors were found to be significantly associated with socioeconomic impact due to COVID-19. These factors include gender, a family member suffering from chronic illness, school-going children, employment status, SES, experienced food scarcity due to COVID-19, and taken additional responsibilities due to the pandemic. The findings from the PPOM analysis indicated that in comparison to female slum dwellers (AOR2 = 0.26; 95% CI [0.12, 0.56]; p = 0.001), male participants had a 3.8 times higher likelihood of experiencing severe socio-economic impact rather than mild and moderate impact due to the pandemic. Additionally, participants whose family members are suffering from chronic illness (AOR1 = 2.09; 95% CI [1.25, 3.51]; p = 0.005) and self-employed participants (AOR1 = 2.46; 95% CI [1.15, 5.26]; p = 0.020) were more likely to experience moderate and severe socioeconomic impact rather than mild impact on COVID-19. Food scarcity during the pandemic was positively and significantly associated with both socioeconomic panels (mild vs. moderate and severe: AOR1 = 2.24; 95% CI [1.37, 3.67]; p = 0.001) and (severe vs. mild and moderate: AOR2 = 4.29; 95% CI [2.38, 7.71]; p < 0.001). However, the socioeconomic impact became more intense when the panel shifted from low to high. The variables ‘age’, ‘marital status’, ‘school-going children’, ‘education level’, ‘smoke tobacco’, ‘taken additional responsibilities due to the pandemic’ satisfied the proportional odds assumption (Brant test value > 0.05). These covariates did not violate the parallel line assumption, and the odds ratio for these variables remains the same (AOR1 = AOR2) across both panels. From the model, In comparison to unemployed participants or housewives, participants who worked as labour or rickshaw puller were 3.2 times (AOR1 = AOR2 = 3.19; 95% CI [1.70, 5.99]; p < 0.001) times more likely to face severe impact during the pandemic. While holding all other variables constant, it is observed that participants who had taken additional responsibilities due to the pandemic were 1.98 times more likely to (AOR1 = AOR2 = 1.98; 95% CI [1.09, 3.62]; p = 0.025) experience severe socioeconomic impact due to COVID-19. After analyzing the data presented in Table 4, it becomes apparent that the impact of COVID-19 on mental health differs between genders. There are several factors associated with the effect of COVID-19 on mental health that vary between males and females. Female slum dwellers aged less than 25 were three times more likely to experience severe mental health problems during the pandemic compared with females aged more than 40 years (AOR1 = AOR2 = 0.26; 95% CI [0.08, 0.75]; p = 0.014) in the slum. Male slum dwellers with a family member suffering from chronic illness (AOR1 = 3.27; 95% CI [1.21, 8.82]; p = 0.019) were more likely to have moderate mental health impact than mild impact during the pandemic compared to those without a chronically ill family member. Female slum dwellers with no formal education were three times more likely to experience severe mental health problems than those who completed primary education (AOR1 = AOR2 = 0.28; 95% CI [0.12, 0.67]; p = 0.004). However, males who completed secondary or higher education (AOR1 = AOR2 = 4.25; 95% CI [1.16, 15.60]; p = 0.029) were more likely to experience severe mental health problems than those without formal education.

Table 4 Partial proportional ordered logistic regression models on the determinants of COVID-19 impact on mental health stratification of gender in the urban slum
Full size table

Females who were suffering from common illness (AOR1 = AOR2 = 2.86; 95% CI [1.25, 6.51]; p = 0.012) and chronic illness (AOR1 = AOR2 = 4.28; 95% CI [1.39, 13.19]; p = 0.011) were more likely to experience severe mental health problems during COVID-19. Male slum dwellers who were smokers (AOR1 = AOR2 = 2.41; 95% CI [1.09, 5.32]; p = 0.030) were two times more likely to experience severe mental health problems due to COVID-19 than non-smokers. Females living in slum areas with no fixed place for garbage disposal were more likely to experience severe mental health impacts due to the pandemic than those having a fixed or city corporation dustbin in the slum (AOR1 = AOR2 = 0.48; 95% CI [0.24, 0.96]; p = 0.040). During the pandemic, males who received support from NGOs or Govt. (AOR2 = 0.12; 95% CI [0.02, 0.57]; p = 0.008) were less likely to experience severe mental health than the mild and moderate impact on mental health due to COVID. Additionally, females who received support (AOR1 = AOR2 = 0.18; 95% CI [0.09, 0.39]; p < 0.001) were less likely to experience mental health problems than those who did not receive support during the pandemic. Compared with females who did not experience food scarcity during the pandemic (AOR2 = 0.22; 95% CI [0.08, 0.61]; p = 0.004), those who experienced food scarcity had a higher likelihood of facing severe mental health problems than the combined categories of mild and moderate impact, even after holding other variables constant. Additionally, males who experienced food scarcity (AOR1 = AOR2 = 0.38; 95% CI [0.17, 0.89]; p = 0.025), lost their jobs or reduced working hours during the pandemic (AOR1 = AOR2 = 4.63; 95% CI [1.83, 11.70]; p = 0.001) were more likely to experience severe mental health problems.Male slum dwellers were more likely to experience mental health challenges due to financial and job-related issues. Men who had lost their jobs due to COVID-19 were four times more likely to struggle with mental health problems during the pandemic.

In a patriarchal society, men are usually responsible for household finances. While most women (68.9%) were unemployed, they relied on the household head's income. However, women who also had financial responsibilities in their households, particularly those working as day laborers or unskilled workers (AOR1 = AOR2 = 2.96; 95% CI [1.04, 8.39]; p = 0.041), were two times more likely to struggle with mental health challenges due to the pandemic than unemployed women. During the pandemic, males who were unable to seek healthcare because of financial difficulties (AOR1 = AOR2 = 3.07; 95% CI [1.22, 7.73]; p = 0.017) were more likely to experience severe mental health problems than those who had no trouble accessing healthcare. In line with this, female slum dwellers who were unable to seek healthcare because of financial difficulties during the pandemic (AOR1 = 4.94; 95% CI [1.49, 16.41]; p = 0.009) were more likely to experience moderate impact on mental health than mild impact due to COVID-19.

4 Discussion

The research paper explores the discernible effects of COVID-19 on slum dwellers in Khulna, a climate-vulnerable city in the southwest region of Bangladesh [27, 29]. Several studies have been done on the slum dwellers residing in this climate-vulnerable area, but none have focused on the socioeconomic and gender-specific mental health impacts of COVID-19. Since many slum dwellers have moved to this city due to natural catastrophes, it is imperative to address the gender-specific impacts during the pandemic. To the best of the authors' knowledge, this is the first quantitative study that demonstrates the gender disparity concerning the socioeconomic and mental health impacts of COVID-19 among slum dwellers in this particular area. This study showed that during the pandemic, male and female slum dwellers experienced socioeconomic and mental health challenges differently. The COVID-19 pandemic has impacted not only the health of people but also their income, where socioeconomically vulnerable people are disproportionately affected by their pre-existing livelihood challenges.

In the representative sample of the KCC slums, we found more female slum dwellers (56.40%) than males (43.60%), while a majority of female slum dwellers (57.9%) experienced food scarcity due to the pandemic, coinciding with a similar study conducted during the pandemic [26]. A higher percentage (55.7%) of male slum dwellers lost their job/ working hours during the pandemic as well as males are more likely to face socioeconomic challenges than female slum dwellers, consistent with a previous study [45]. Whereas (44.3%) of males reported their financial situation and employment status remained unchanged, a possible reason could be those who worked in a jute mill during the pandemic had a permanent earning source. However, since the closure of the jute mill factories, many of these men (from ward no 8 and 11) have reported losing their jobs and are currently struggling even more than during the lockdown period. Furthermore, the socioeconomic impact was found most notable in unskilled workers or day laborers, while the majority of female slum dwellers reported being housewives. These findings align with previous research [27] conducted in Khulna slum areas during the pandemic, which reported that most women were found to be unemployed due to the lack of skills relevant to their jobs (tailoring, handicrafts, etc.). Furthermore, we found that slum dwellers who smoke tobacco faced severe socioeconomic impacts during the pandemic. This finding aligns with a recent longitudinal study [46], which reported that COVID-19 economic hardship was associated with increased tobacco smoking. Increased financial hardship may lead to an increase in smoking behaviors, and higher costs connected with purchasing tobacco products may put additional financial strain on households. Slum dwellers who had a family member with chronic illness were found two times more likely to experience moderate and severe socioeconomic hardship during the pandemic, as a previous study reported that the presence of chronic illness in a household is a key determinant of high OOP (Out-of-Pocket) payments and financial hardship [47]. Socioeconomic hardship during the pandemic associated with employment status and food scarcity due to COVID-19, echoes earlier findings [22].

The quantitative study revealed notable disparities between the severity of mental health issues among male and female slum dwellers. Specifically, 23.7% of female slum dwellers reported experiencing more severe mental health challenges during COVID-19, as compared to 9.7% of males. Previous studies [48, 49] conducted during the pandemic showed similar findings that women experienced higher rates of mental health symptoms like anxiety, depression, stress, and insomnia. This gender disparity in mental illness can be attributed to various factors, including biological, hormonal, social, and cultural influences [49, 50]. According to research, women may be more susceptible to anxiety disorders than men due to monthly and life-span hormonal fluctuations (notably progesterone and oestradiol hormones) [51, 52]. Moreover, various sociocultural variables, such as low self-esteem and emotional expressiveness, helplessness from the burden of caregiving, and limited engagement in socioeconomic and decision-making activities, may also impact women's psychological health [53, 54]. Moreover, the mental health of slum dwellers during COVID-19 was associated with several factors, such as individuals with a ‘chronically ill family member’, ‘received support from NGO or Govt. during the pandemic’, ‘experienced food scarcity’, ‘difficulty accessing basic needs’, ‘unable to seek healthcare due to financial difficulties’, and ‘borrowed money to meet daily needs during COVID-19’, align with earlier studies [55,56,57]. Severe mental health issues found in men living with a chronically ill family member may be because men are less used to providing care and spending time with family. We found younger (≤ 25 years) female slum dwellers were more likely to experience severe mental health issues due to the pandemic than over 40-year-old females. Earlier studies found similar results and also reported possible reason could be younger individuals have experienced more drastic changes in their daily routines due to the lockdown measures than older individuals [58, 59]. In addition, we found that females with lower household income (less than 7 K Taka) and no formal education reported severe mental health issues during the pandemic, consistent with a previous study [22]. However, male slum dwellers with secondary or higher education experienced more severe mental health problems during the pandemic, which aligns with a previous study [60] conducted during the outbreak of COVID-19. A possible explanation could be that during the pandemic family members tended to rely on educated male slum dwellers to provide for the family. However, even the educated ones lost their jobs, leaving those male slum dwellers more time to stress on other aspects of life like work status, family health, and household income. The study revealed that female slum dwellers with chronic illnesses and a lack of a fixed place for garbage disposal within their community experienced severe mental health issues, which coincides with previous studies [22, 61, 62]. Slums with no fixed garbage disposal create an unhealthy environment for residents to live in, which can affect their mental health. Nevertheless, no prior studies were conducted particularly for the targeted group of females’ mental health and their surrounding environmental conditions. In many South Asian countries, improper disposal and landfilling practices pollute the environment by emitting unpleasant odors into the atmosphere, which worsen during the rainy season [63, 64]. Notably, a lack of proper incineration facilities or equipment can increase the volume of ash residue and cause air pollution, which poses risks to human health [63]. Previous research revealed that living near a landfill can impact psychological well-being, and acidic gas emissions from landfills can significantly affect human health [64, 65]. Moreover, a previous study [66] agrees with our findings that the sanitation facilities during COVID-19 affected the mental health of women living in slum areas. A possible explanation is that inadequate sanitation facilities can cause considerable hygiene hazards, particularly for female slum dwellers [27], and the lack of privacy in shared or unimproved sanitation facilities can lead to anxiety and emotional distress [67]. Male slum dwellers who smoke tobacco and lost jobs during COVID-19 reported severe mental health problems due to the pandemic, findings align with previous studies [22, 59] that mentioned smoking and joblessness associated with depressive symptoms among slum dwellers. During the COVID-19 pandemic, men and women who did not receive support from NGOs and the government and were unable to seek medical care due to financial difficulties reported mental health issues, consistent with previous studies[68,69,70]. However, a previous study indicated that receiving unequal food aid is associated with insomnia among slum dwellers [49], highlighting the need for fair food distribution during any crisis. From our findings, it is evident that males and females faced distinct mental health challenges during COVID-19. A recent study reported that women and young adults experienced significant psychological distress related to the pandemic [71]. The study's findings are valuable for government and community policymakers to build a sustainable community. However, there are some limitations to consider. The chosen model in our study (PPO model) is less straightforward to interpret than the ordinal logistic regression model. Moreover, we conducted the research in a South Asian city with a distinct social and economic framework, so the study's findings may not be applicable to other cities worldwide. The survey took place over a specific period, and the mental health status of individuals in the slum area may continue to alter over time. We recommend further longitudinal studies to find more in-depth relationships between certain factors..

5 Conclusion

This research addresses the gender-specific effects of the COVID-19 pandemic and provides a comprehensive understanding of how it impacted individuals within their respective socioeconomic and mental health contexts. In times of crisis, such as pandemics, it is quite common for male slum dwellers to experience a job loss. In certain situations, families rely solely on female family members as earners. Community leaders, government, and non-governmental organizations should implement skills-based or vocational training for men while also providing training to women in tailoring and handcrafts to address this issue. Slum dwellers can better navigate challenging times and ensure a stable source of income through skill-building training.

Additionally, it is crucial to prioritize all individuals' mental health and well-being, especially women who may face additional challenges in slum areas without proper sanitation facilities. In light of the COVID-19 pandemic, having a fixed place for garbage disposal is crucial to maintain cleanliness and prevent any health hazards. Providing a safe and sound environment is essential for human development, and improved sanitation facilities should be a top priority. Specifically, ensuring women have access to gender-segregated improved sanitation facilities and having a fixed garbage disposal in the slum areas can make a big difference. In order to address mental health concerns related to the pandemic, fair and adequate food and financial aid should be provided during times of crisis. At the community level, awareness programs should strive to disprove stereotypes and reduce stigma. Awareness regarding immediate support and counseling should be promoted during any pandemic, emphasizing that mental health issues are just as valid as physical health concerns and deserve equal attention.