Trend and gender-based association of the Bangladeshi student suicide during the COVID-19 pandemic: a GIS-based nationwide distribution

Abstract

Background:

Students are one of the most vulnerable groups to suicide. Before the COVID-19 pandemic, a Bangladeshi study was conducted assessing their suicide patterns regarding gender-based associations. But how has the pandemic changed the Bangladeshi students’ suicide patterns is not studied yet, which is investigated herein. Besides, for the first time, this study provides GIS-based distribution of suicide cases across the country’s administrative district.

Methods:

As Bangladesh has no suicide surveillance system, this study utilized media reporting suicide cases following the prior studies. A total of 127 students’ suicide cases from March 2020 to March 2021 were finally analyzed after eliminating the duplicate ones, and data were synthesized following the prior studies. Arc-GIS was also used to distribute the suicide cases across the administrative district.

Results:

Results revealed that female (72.4%; n = 92/127) was more prone to die by suicide than males. About 42.5% of the cases were aged between 14 and 18 years (mean age 16.44 ± 3.512 years). The most common method of suicide was hanging (79.5%; n = 101), whereas relationship complexities (15.7%), being emotional (12.6%), not getting the desired one (11%), conflict with a family member (9.4%), academic failure (9.4%), mental health problem (8.7%), sexual complexities (6.3%), scolded or forbidden by parents (3.9%) were the prominent suicide causalities. In respect to gender and suicide patterns, only the suicide stressor was significantly distributed, whereas the method of suicide was significantly associated with GIS-based distribution. However, a higher number of suicide cases was documented in the capital (i.e. Dhaka) and the northern region than in its surrounding districts.

Conclusions:

The findings reported herein are assumed to be helpful to identify the gender-based suicide patterns and suicide-prone regions in the time of the COVID-19 pandemic to initiate suicide prevention programs of the risky students.

Introduction

After the COVID-19 pandemic inception, students’ normal lives have been disrupted because of its stressful situation. The typical schooling has been converted into the online mode, which can be unfavorable to many students (al Mamun et al., 2021; Pragholapati, 2020). Mental health problems like depression, anxiety, and stress are intensified among the students by the issues related to the pandemic such as fear of COVID-19, panic toward it, worry about it, a social distancing that impaired social relationships, movement restrictions, feeling of helplessness, amplified loneliness, nervousness, and so forth (al Mamun et al., 2021). Similarly, these stressors are being reported to intensify suicidal events in extreme cases (Lathabhavan & Griffiths, 2020; Manzar et al., 2021; Tanaka & Okamoto, 2021). Similar to the prior contagious outbreaks such as the 1918 influenza pandemic, 2003 SARS epidemic, etc., suicide rate increment during the COVID-19 pandemic is claimed (Calderon-Anyosa & Kaufman, 2021; Kahil et al., 2021), although recent evidence from five upper-middle-income and 16 high-income countries found there is no increment of suicide after the COVID-19 pandemic inception (Pirkis et al., 2021).

Suicide is a multifactorial public health problem for all age groups across the globe. Yearly, more than 800,000 people pass away because of it (WHO, 2018). About 80% of the global suicides occur in low- and middle-income countries like Bangladesh (WHO, 2018), where 10,000 people are said to die by suicide each year (Mashreky et al., 2013; WHO, 2019). The WHO estimated youth and adolescents as the most vulnerable group to suicide (WHO, 2018); correspondingly, about 61% of the Bangladeshi suicide cases had been reported to less than 30 years as per a media reporting suicide study (Shah et al., 2017). However, as reported by a systematic review, the global student suicide attributing factors included hostile and stressful academic and educational environment, relationship complexities, economic crisis, financial distress, alcohol, and drug use, family and parental problems, traumatic events exposure, psychological problems, etc. (Yozwiak et al., 2012).

In Bangladesh, the first COVID-19-related suicide case was reported on 25 March 2020 (Mamun & Griffiths, 2020), whereas on 11 June 2020, the first student suicide case was reported (Mamun, Chandrima, et al., 2020). Online schooling issue (i.e. the victim was not interested to take part in the online exam, which turned into familial conflicts) was alleged as the suicide attributing factor (Mamun, Chandrima, et al., 2020). Similarly, an educationally gifted 15-year old Indian girl died by suicide because of a lack of resources to attend online classes due to extreme financial poverty in their family (Lathabhavan & Griffiths, 2020). During the initial period of the pandemic, Manzar et al. (2021) observed a total of 37-student suicide cases from 11 countries, whereas diverse suicide stressors were reported. Although a few Bangladeshi case-series studies assessed the general people’s suicide during the pandemic (Bhuiyan et al., 2020; Boshra et al., 2020), none considered the student cohort, which is remedy herein for the first time. See Mamun (2021) for a systematic review on Bangladeshi suicide and suicidal behaviors related to the COVID-19 pandemic.

Heterogeneity in gender-based suicide was found in the Bangladeshi general population; females were found more dying as reported in the historical data (Arafat, 2019). The gender-based trend of suicide is little known in Bangladesh, although utilizing the press media reporting suicide cases of the Bangladeshi general student suicide, a higher portion of the male student suicide occurrences was reported in a study (71.4%) (Mamun, Siddique, et al., 2020). However, how the COVID-19 pandemic has changed the general student suicide patterns regarding its gender-based distribution can be worthy of investigation, which is also focused on the present study. In addition, Bangladesh lacks available data on suicide because of there is no national surveillance system and nationwide suicide study (Arafat, 2019; Mamun, Siddique, et al., 2020). Region-based suicide mortality data is impossibly retrievable for any population group of the country because there is no previous estimation. Therefore, for the first time, this study attempts an insight on GIS-based distribution of the Bangladeshi student suicide cases reported in the Bangladeshi press media.

Finally, the present study provides information on 1-year long-press media reporting suicide cases of the Bangladeshi students during the COVID-19 pandemic, where the gender-based difference is also observed in consideration of comparing the findings with a study that was conducted with similar methods among the same cohort during the normal period (Mamun, Siddique, et al., 2020). In addition, district-wise distribution of the student suicide cases is also provided, which can be helpful for better understanding of the heterogeneity of the risky districts for implementing proper policy.

Methods

Study design, data source, and ethics

The present study is retrospective and collected the Bangladeshi student suicide data from the press media reports. As there is no nationally representative suicide database or surveillance system, alternative sources (e.g. press media reports) for retrieving suicide data are considered in this study (Arafat, 2019; Shah et al., 2017). Thus, suicide cases reported in the press media are mostly utilized for retrospective studies in Bangladesh to gather the victims’ basic information (Mamun, Siddique, et al., 2020; Shah et al., 2017). This method is also used in South Asian countries with a lack of national suicide databases, which became an emergence source of suicide study during the COVID-19 pandemic for rapid assessment of suicide information (e.g. India Panigrahi et al., 2021; Sripad et al., 2021), Pakistan (Mamun & Ullah, 2020, etc.). As the data were collected from the secondary source and used anonymously in this study, no ethical approval or consent to participation is required for this study.

Data collection procedure

Suicide data concerning the Bangladeshi students were collected by using online news media platforms. A systematic search was carried out in the Google search engine, utilizing the relevant keywords in Bangla and English languages. Search keywords include ‘suicide’, ‘suicide news’, ‘suicide attempt’, ‘self-harm’, ‘student’, ‘school student’, ‘college student’, ‘university student’, ‘adolescent’, ‘youth’, ‘COVID-19’, ‘pandemic’, ‘outbreak’, ‘2020’, and ‘2021’. Student suicide data of the past 15 months (January 2020–March 2021) reported in the Bangladeshi press media were collected in this study. It should be noted that data from January 2020 to March 2021 were only used for Figure 1a, with data from March 2020 to March 2021 being used for other analyses and Figures. However, after eliminating the duplicate cases, a total of 127 student suicide cases were used for formal analyses in this study.

Figure 1.

(a) Student suicide tendency based on the cases from January 2020 to March 2021, (b) nationwide distribution of the student suicide cases, (c) gender-based distribution of the suicide cases, (d) distribution of the suicide cases’ mean age, (e) educational grade-based distribution of the suicide cases, and (f) distribution of the methods of the suicide cases.

Note.Figure 1a includes suicide cases from January 2020 to March 2021, although the rest figures are based on the suicide cases after the COVID-19 pandemic inception in Bangladesh (i.e. from March 2020 to March 2021).

Data presentation

First of all, the extracted data were gathered in a Microsoft Excel file. Then, for conducting the formal statistical analysis, media reports were analyzed and categorized to vanish the inconsistencies of the data. Then, for synthesizing the suicide reports, the prior study utilizing the same methods within the same cohort was followed (Mamun, Siddique, et al., 2020), where the lead author (MAM) expert in media reporting suicide took the lead. Finally, data were reported based on the suicide reporting date, gender, age, district, educational level, method of suicide, and suicide stressors. However, a short description of the synthesized variables is given below.

Age group

The age of the victims was collected from the media reports, and it was categorized based on 5-year intervals to conduct a formal analysis as followed by the prior student suicide study (Mamun, Siddique, et al., 2020). The mean age of the reported cases was 16.44 (±3.51) years.

Educational level

The education level was ranked based on the existing Bangladeshi education system; also it was followed the prior study (Mamun, Siddique, et al., 2020). This includes primary, secondary, higher secondary, and undergraduate educational levels of the suicide victims who studied in grades 1 to 5, 6 to 10, 11 and 12, and more than 12, respectively.

Methods of suicide

Following the literature, the method of suicide was categorized into three types: hanging, poisoning, and others (that includes firing up the body with kerosene oil, cutting the throat, taking gastric tablets, and jumping from the terrace).

Suicide stressors

The suicide stressors were categorized into a few major types, including (i) being emotional, (ii) scolded or forbidden by parents, (iii) not getting the desired one, (iv) relationship complexities, (v) sexual complexities, (vi) conflict with family members, (vii) mental health problems, (viii) academic failure, and (ix) others. Note that some cases reported multiple suicide reasons, where the most affecting one was kept in this study. For example, ‘scolded or forbidden by parents’ includes issues like forbidding to play a video game, going out during raining, panicking, or being scolded by a parent. Not getting a motorbike, festival dress, smartphone, new dress, etc., were taken into ‘not getting the desired one’. For categorizing the sexual complexities, being raped, being harassed (eve-teasing), being blackmailed by the edited naked picture updated on social media were considered. Conflicts with family members (mother, father, brother, and sister) were included within conflict with family members. Stress, depression, mental, and social pressure were included in the mental health problems. Lastly, kidnapping or frustration or mother and sister being assaulted or forceful marriage or cheated or limitation of movement or blackmailing and threat or false accusation and threat or feeling ashamed or financial problem were considered other factors for dying by suicide.

Statistical analysis

Data analysis were performed using Microsoft Excel 2019 and Statistical for Social Science (SPSS) version 25. Descriptive (frequency, percentage, mean, and SD) and inferential statistics (chi-square) were executed in this study with a p-value <.05 as statistically significant. Besides, ArcGIS 10.8 was also used to present the district-wise heterogeneity of suicide cases, gender, age group, education level, and suicide methods. It should be noted that data from January 2020 to March 2021 were used for only Figure 1a, although other analyses were performed with the suicide cases related to the COVID-19 pandemic. Note that the pandemic started in March 2020 in Bangladesh, whereas March 2020 to March 2021 suicide cases were related to the pandemic (N = 127).

Results

Distribution of the student suicide cases

Table 1 demonstrates that most suicide victims were female (n = 92/127, 72.4%). About 42.5% of the cases were aged between 14 and 18 years, whereas the mean age of the victims was reported as 16.44 (±3.512) years (age range 9–23 years). About 43.3% were students of secondary level, where the second most suicide-prone group was found in higher secondary level (26%). Most of the victims preferred to die by hanging (79.5%), whereas 10.2% died by poisoning as a suicide method. Relationship complexities were reported as the prominent suicide causative factor leading to 15.7% student suicidality; whereas other factors include (i) being emotional (12.6%), (ii) not getting the desired one (11%), (iii) conflict with family members (9.4%), (iv) academic failure (9.4%), (v) suffering from mental problems (8.7%), (vi) sexual complexities (6.3%), (vii) scolded or forbidden by parents (3.9%), etc. However, considering the gender-based distribution of the student suicide, the suicide stressor was only significantly associated (χ² = 16.204, p = .04).

Table 1.

Gender-based characteristics distribution of the student suicide.

Variables	Total (n, %)	Gender		Statistics
		Male (35, 27.6%)	Female (92, 72.4%)	χ2 test value	p-value
Age group (mean, SD = 16.44, 3.512)
9–13 years	18, 14.2	5, 18.5	13, 18.1	1.995	.36
14–18 years	54, 42.5	12, 44.4	42, 58.3
19–23 years	27, 21.3	10, 37	17, 23.6
Education level
Primary	7, 5.5	2, 6.3	5, 5.6	5.277	.15
Secondary	55, 43.3	9, 28.1	46, 51.1
Higher secondary	33, 26	11, 34.4	22, 24.4
Undergraduate	27, 21.3	10, 31.3	17, 18.9
Suicide methods
Hanging	101, 79.5	29, 93.5	72, 81.8	2.741	.25
Poisoning	13, 10.2	1, 3.2	12, 13.6
Others	5, 3.9	1, 3.2	4, 4.5
Suicide stressors
Being emotional	16, 12.6	3, 9.4	13, 17.6	16.204	.04
Scolded or forbidden by parents	5, 3.9	2, 6.3	3, 4.1
Not getting the desired one	14, 11	7, 21.9	7, 9.5
Relationship complexities	20, 15.7	6, 18.8	14, 18.9
Sexual complexities	8, 6.3	0	8, 10.8
Conflict with family members	12, 9.4	1, 3.1	11, 14.9
Academic failure	12, 9.4	3, 9.4	9, 12.2
Mental problem	11, 8.7	7, 21.9	4, 5.4
Others	8, 6.3	3, 5	5, 6.8

The tendency of the student suicide cases

Figure 1a demonstrates the student suicide trend based on the cases from January 2020 to March 2021. There had been ups and downs in suicide cases in the last 15-months. A gradual increment of suicide cases from January 2020 to June 2020 was observed, whereas April was exceptional. Likewise, more suicide cases were recorded from August 2020 to December 2020, except in November, where a smaller number of cases were reported than that month. In 2021, the same number of cases were documented in January and March, whereas fewer cases were noted in February.

Nationwide distribution of the student suicide cases

Figure 1b displays the distribution of the student suicide cases across all the administrative districts of Bangladesh. Dhaka, the country’s capital, documented most likely to have a higher number of suicide cases than its surrounding districts. Similarly, a higher number of suicide cases was observed in the northern districts such as Bogra, Sirajgonj, Mymensingh, and Netrakona. In addition, Jhenaidah had reported a higher number of suicide cases.

Gender-based distribution of the student suicide cases

Figure 1c illustrates the gender-based distribution of the suicide across districts; this association was not statistically significant (χ² = 60.535, p = .074). For males, a higher number of suicide cases were reported in Munshigonj and Jaipurhat. In contrast, Mushingonj, Mymensingh, Tangail, Bogra, Rajshahi, Jhenaidah, Gazipur, Dinajpur, Noakhali, Sirajgonj, and Dhaka districts were documented as higher suicide cases for the female genders.

Distribution of the student suicide cases’ age

Figure 1d shows the distribution of the suicide cases means age across districts. The association was not statistically significant (F = 0.942, p = .578). The figure depicted a higher mean age of the suicide victims in Tangail, Mymensingh, Chittagong, Noakhali, Kustia, Kurigram, etc.

Education-based distribution of the student suicide cases

Figure 1e represents the educational grade-based distribution of the suicide cases’ educational grades across districts. That did not show any statistically significant relationship (F = 0.642, p = .892). The districts of the northern region (i.e. Thakurgaon, Kurigram, Jaipurhat, Gaibanda); southern part (i.e. Comilla, Chandpur, Shariatpur); and some districts of the western part of the country had reported the suicide victims with a higher educational grade.

Distribution of the student suicide cases’ methods

Figure 1f presents the distribution of the suicide methods across districts that were statistically significant (χ² = 120.564, p = .034). As aforementioned, the majority of the victims preferred hanging as the suicide method, whereas few cases were died either by poisoning or by other methods. The hanging method was found among the districts, including Thakurgaon, Bogra, Sirajgonj, Munshigonj, Jhenaidah, Dhaka, etc. Poisoning and other methods of suicide were described in the district of Mymensingh, Tangail, Hobiganj, etc.

Discussion

To the best of the authors’ knowledge, the present study, for the first time, investigated the student suicide cases during the COVID-19 pandemic in Bangladesh, where gender-based and GIS-based distribution of the suicide characteristics are presented. It is anticipated that the present findings would be beneficial for the authority to alleviate unwanted suicide occurrences by introducing preventive approaches.

The number of student suicide cases after one year of the COVID-19 pandemic was 127 in this study, but a previous study utilizing the similar press reporting suicide method during the normal period gathered data over a period of 16 months (from January 2018 to June 2019) found 56-cases (Mamun, Siddique, et al., 2020); this appears an increment of the Bangladeshi student’s suicide. During and aftermath of an outbreak like the COVID-19, suicide rate increment is not unusual (Calderon-Anyosa & Kaufman, 2021; Kahil et al., 2021). Although Pirkis et al. (2021) based on the suicide surveillance data from 21-upper-middle-income and high-income countries, estimated monthly suicide incidences using an interrupted time-series analysis and compared the trend of suicide before and during the pandemic, there was no increment of suicide. Therefore, the present study’s findings from Bangladesh may not necessarily reflect those findings reported from developed countries due to methodological disparities as well as the nature of suicide under the stressful situation.

Suicide among the female students was prevalently observed (72.4% female vs. 27.6% male) in this study, representing the gender-based suicide heterogeneity in Bangladesh. That is, the historical data suggest that suicide in Bangladeshi general females is higher than the male ones (Arafat, 2019). However, this gender-based finding totally contrasts with the global suicide trend, where males are being reported more dying by suicide (i.e. 13.7 males vs. 7.5 females per 100,000 population) (WHO, 2019). Moreover, compared with Mamun, Siddique, et al.’s (2020) study among the student suicide during the normal period, male students were found dying more (71.4% male vs. 28.6% female), suggesting pandemic has impacted the gender-based suicide trend. Apparently, this gender-based finding let allow to think about considering the female student mental health, although the male gender cannot be ignored at all. Besides, additional studies are suggested for identifying the underlying suicide factors across genders.

About 42.5% of the students aged 14 to 18 years died by suicide; similarly, secondary school-going students (e.g. grades 6–10) were found accounting for suicide in 43.3% of the present cases. This departs from the prior study, where about 66.1% of Bangladeshi student suicide belonged to tertiary education level (undergraduate and above) (Mamun, Siddique, et al., 2020). However, the educational level was not a significant factor influencing gender-based suicide distribution in this study, although it was in the prior study (Mamun, Siddique, et al., 2020). In addition, the methods used for suicide generally depend on availability. Whereas hanging is predominately observed as the most-used suicide method in Bangladesh, followed by poisoning, ingestion of insecticides, firearms, jumping in front of a train, jumping from heights, etc. (Arafat, 2017; Ferdous & Alam, 2021). About 79.5% of the victims hanged as per this study, whereas the rate was almost the same in the prior student suicide study (75.0%) (Mamun, Siddique, et al., 2020). In addition, all but one female suicide was reported by hanging, which was significant in terms of gender-based distribution in the prior study (Mamun, Siddique, et al., 2020), but such suicide method-wise association with gender was not significant in this study.

Suicide is a phenomenon of multifactorial, which is completed by the combination of multiple factors (e.g. neurobiology, personal, and familiar history, stressful life events, socio-cultural environment, emotional dysregulation, and so on). In this study, 15.7% of the suicide victims were reported dying as of relationship complexities, whereas other stressors included being emotional (12.6%), not getting the desired one (11%), conflict with family members (9.4%), academic failure (9.4%), suffering from mental problems (8.7%), sexual complexities (6.3%), scolded or forbidden by parents (3.9%), etc. Although the relationship problem as a suicide stressor is consistent with the prior Bangladeshi study, where it was the prominent one in this study (19.6%), other factors were included to be academic distress and failure, quarreling with parents, parental problems, economic distress, and mental health problems (Mamun, Siddique, et al., 2020). Both two studies found a gender-based significant association of suicide stressors. During the COVID-19 pandemic, a study focusing on student suicide causalities found most of the stressors were related to mental health problems like depression, loneliness, psychological distress, etc., along with online schooling issues, overwhelming academic distress, TikTok addiction-related psychological distress, being infected with COVID-19 (Manzar et al., 2021). In addition, students could not go outside because of strict restrictions, which exacerbated their psychological and emotional distress (al Mamun et al., 2021). Negative moods such as intense fear, anxiety, mental instability, and economic distress experienced during the outbreak lead to decreased hope and increased suicidal behaviors, including actual suicide completions (Calderon-Anyosa & Kaufman, 2021; Kahil et al., 2021; Mamun, 2021). This may affect the identified stressors of suicide in this study; however, we recommend further studies explaining the actual pathway of suicide to this cohort.

With the limited available information, the Jhenaidah District, one of the 64 districts, was reported to be one of the highest suicide-prone regions in the country (Khan et al., 2020). GIS-based distribution based on the heterogeneity of suicide cases, gender, victim age, educational grade, and reason for suicide was evaluated across the district that was never explored in the Bangladeshi context (Figure 1). Despite being the main city of Bangladesh, Dhaka documented most of the student suicide cases compared to other districts. Males from the district of Munshiganj and Jaipurhat had a higher number of suicide cases, whereas it was the northern region for the suicide of the female gender. The higher mean age of the suicide victims was found across the diverse district in the country. In addition, suicide victims’ higher educational grades and preferable methods of suicide were visualized in the northern part. Therefore, a suicide prevention strategy is needed for the most vulnerable region in the country, where the present GIS-based distribution can be worthy of use.

Due to the scarcity of nationwide suicide surveillance data in Bangladesh, the present study analyzed data derived from the newspaper, limiting our study as newspaper data may not explore a suicide case from different aspects. In addition, many suicide cases could not be reported in the press media (e.g. remote areas suicide cases due to the lack of reporters), and some of the reported cases could not be included in this study (as they did not meet the case inclusion criteria), limiting the present findings (Mamun, Siddique, et al., 2020). However, despite these limitations, this study provides information on (i) student suicide during the COVID-19, (ii) trend of suicide compared to the normal time, (iii) compared the findings with the prior study conducted within the same cohort, and (iv) provided nationwide GIS-based distribution of suicide cases.

Concluding remarks

The overall scenario of the student suicide demands an immediate response from the authority of Bangladesh as no supportive governmental center exists in the country, which helps against such types of issues that lead to suicidal behavior. Therefore, implementation of suicide prevention strategies is required with a special focus on students of all ages, including establishing awareness programs against suicide during this COVID-19 crisis moment, consistently providing mental health care during online classes in such a stressful situation, the COVID-19 pandemic provides.