Social Determinants of Suicidal Ideation among Adolescents in Rural Bangladesh

Social Determinants of Suicidal Ideation

among Adolescents in Rural Bangladesh

Afroza Begum

Main supervisor: Gloria Macassa, Professor

Co-supervisors: Joaquim Soares, Professor, AKM Fazlur Rahman,

Professor, Eija Viitasara, Associate Professor, Hamid Reza Khankeh

Professor

Faculty of Human Sciences

Thesis for Licentiate degree in Health Sciences

Mid Sweden University

Akademisk avhandling som med tillstånd av Mittuniversitetet i Sundsvall

framläggs till offentlig granskning för avläggande av filosofie licentiatexamen

fredagen den 26 februari 2021, klockan 10:00 via Zoom, Mittuniversitetet,

Sundsvall. Seminariet kommer att hållas på engelska.

Social Determinants of Suicidal Ideation among

Adolescents in Rural Bangladesh

© Afroza Begum, 2021-02-26

Printed by Mid Sweden University, Sundsvall ISSN: 1652-8948

ISBN: 978-91-88947-90-1 Faculty of Human Sciences

Mid Sweden University, SE-851 70 Sundsvall Phone: +46 (0)10 142 80 00

To my beloved parents

To my beloved daughters, Tasneem and Tashfia

To my beloved husband

And

Acknowledgement

I feel deeply indebted to Almighty Allah and a great many people whose inspiration and support enabled me to complete my thesis.

In particular, I would like to thank my main supervisor Professor Gloria Macassa, for her invaluable guidance, encouragement, academic inspiration and generous help. From the inception of this thesis to its completion, she has contributed so much to my studies. She was always there for me whenever I needed assistance and suggestions throughout my writing. I gained not only the knowledge, but the rigorous scientific approach and spirit of dedication to the work that needed to be done. Her steady, reliable supervision enabled me to complete this challenging project.

Professor Joaquim Soares, my co-supervisor, for giving me practical guidance and encouragement in my study.

Professor AKM Fazlur Rahman, my co-supervisor, for his overwhelming support in collecting data on suicidal ideation as well as for sharing his immense knowledge, scientific assistance and comments in my licentiate studies. I will always remember his encouragement to keep writing and moving forward.

Associate Professor Eija Viitasara, my co-supervisor, who always had words of encouragement and enthusiastically offered assistance in any way she could throughout the research project.

Professor Hamidreza Khankeh, University of Social Welfare and Rehabilitation Science, my co-supervisor, a true personality, his patient support for my study plan encouraged me to continue the studies.

Professor Sadiqa Tahera Khanam, former director and professor at MCH, NIPSOM and my MPH supervisor, for her support during different stages of my thesis. She has always been extremely accommodating whenever I had a question about my research or writing.

Professor Shamsun Nahar, professor of community medicine, King Khaled University, Abha, Saudi Arabia and Dr. Meera Momtaz Sabeka, consultant neurologist, East Kent Hospital, UK, my best friends and ex-colleagues who inspired me and helped me throughout my PhD studies.

Professor Jahanara Begum and Professor Mahmudul Haque, my dear ex-colleagues at NIPSOM, for their tremendous assistance while completing my studies.

Dr. Aminur Rahman, director, Dr. Saidur Rahman Mashreky, director, Md. Al-Amin Bhuiyan, deputy team leader and Md. Abu Talab, senior statistician at CIPRB, for all their support during my studies.

The data collectors and respondents whose efforts made data collection successful on this sensitive issue.

Last but not least, I owe so much to my dear parents, Md. Abul Hossain and Rizia Khanom. Their unwavering love and confidence in me have encouraged me to keep moving forward in my study and career. I express my gratitude to all my beloved family members, my husband H. M. Akhter Khan, daughters Takfi Tasneem, Tashfia Kawakib, my brother Iqbal Hossain and sister-in-law Nupur for their understanding, support and encouragement to finish my studies.

Table of Contents

Abstract ... ix

List of Papers ... xi

List of Abbreviations ... xii

1. Introduction ... 1

2. Background ... 2

2.1 Suicidal Ideation Worldwide ... 2

2.2. Suicidal Ideation in Developing Countries and Asia ... 3

2.3. Suicidal Ideation in Bangladesh ... 5

3. Determinants of Suicidal Ideation among Adolescents ... 7

3.1 Biological and Psychological Determinants ... 7

Biological Risk Factors ... 7

Psychological Risk Factors ... 8

3.2. Environmental Determinants ... 9 3.3. Socio-Economic Determinants ... 10 4. Conceptual Model ... 11 5. Rationale ... 11 6. Objectives ... 13 6.1 General Objective: ... 13 6.2. Specific Objectives ... 13 7. Method ... 13

7.1. Setting and Participants ... 13

7.2. Study Sample and Design ... 14

7.3. Sampling Procedure ... 14 7.4. Survey Procedure ... 15 7.5. Measurement of Variables ... 16 7.5.1. Dependent Variable ... 16 7.5.2. Independent Variables ... 16 7.6. Statistical Analyses ... 16

7.7. Ethical Considerations ... 17

8. Results ... 18

8.1. Study 1 ... 18

8.2. Study 2 ... 20

9. Discussion ... 24

9.1 Prevalence of Suicidal Ideation among Adolescents and Young Adults in Rural Bangladesh ... 24

9.2. Parental Socio-Economic Status and Suicidal Ideation among Adolescents in Rural Bangladesh ... 25

9.3. Strengths and Limitations ... 26

9.4. Future Research ... 26

10. Conclusions ... 27

ix

Abstract

Background and Objective: Suicide is a leading cause of death and a critical public health problem among adolescents worldwide. However, adolescent suicidal behavior is a neglected public health issue, especially in low- and middle-income countries, such as Bangladesh. There is evidence from developed countries showing that suicidal ideation (SI) among adolescents is related to social indicators as well as individual risk factors. Scarce evidence from studies in low- and middle-income countries suggests that social factors may influence adolescent SI. In Bangladesh, only a few studies have focused on the social determinants of suicide, and the present study is the first to examine the prevalence and social determinants of suicidal ideation among adolescents in rural Bangladesh. Therefore, the present thesis assessed the effects of socioeconomic status (SES) on adolescent SI.

Methods: The present thesis is based on two studies, which used data from a cross-sectional survey conducted in 2013 among 2,476 adolescents aged 14-19 years selected randomly in a rural community in Bangladesh. An adapted version of the WHO/SUPRE-MISS questionnaire was used to collect data in the Raiganj sub-district, an area surveyed within the Injury Prevention Program of the Centre for Injury Prevention and Research, Bangladesh (CIPRB). Study I estimated the prevalence of suicidal ideation among adolescents and investigated what factors were associated with suicidal ideation. Study II examined the relationship between parental socio-economic factors and suicidal thoughts among adolescents.

Results: Study I showed that lifetime prevalence of suicidal ideation was 5% among adolescents. The majority of adolescents with suicidal ideation were female (67, 5.3%), unmarried (106, 5.5%) and students (86, 6.2%). Suicidal ideation was significantly associated with age, education, occupation and living with parents or others. Respondents aged 18-19 years, had a Secondary school certificate (SSC) and Higher Secondary school certificate (HSC) or higher, were day laborers, living without parent had odds ratios of 2.31 (CI 1.46-3.65), 2.38 (CI 1.51-3.77), 4.15 (CI 2.41-7.14), .28 (CI 0.13-0.60) and 1.80 (CI 1.07-3.03), respectively.

In study 2, results demonstrated that suicidal ideation was more common among adolescents with parents from the low-income group (5.5%). In addition, adolescent SI was found to be significantly associated with parental education, marital status and house ownership. Adolescents whose parents received an SSC or higher had an SI odds ratio of 2.10 (CI 1.21-3.64) and 1.92 (CI 1.15-3.23) for mothers and fathers respectively. Suicidal ideation among adolescents with single parents was higher with an odds ratio of 3.00 (CI 1.75-5.19) in comparison to adolescents who had both parents. Adolescents whose parents owned a house had an odds ratio of 0.14 (0.05, 0.35).

x

Conclusion: The thesis found that the prevalence of lifetime suicidal ideation among adolescents is moderately high in a rural community of Bangladesh. Furthermore, it was observed that personal characteristics, parental marital status, education and home ownership were statistically significantly associated with suicidal ideation among adolescents even after adjusting for potential confounders.

Keywords: Prevalence, Factors, Socio-Economic, Parents, Suicidal Ideation, Rural, Adolescents

xi

List of Papers

This thesis is based on the following papers:

Paper I: Begum A, Rahman, F, Rahman A, Soares J, Khankeh HR & Macassa

G. Prevalence of suicidal ideation among adolescents and young adults in

rural Bangladesh. International Journal of Mental Health 2017;

DOI:10.1080/00207411.2017.1304074.

Paper II: Begum A, Viitasara E, Soares J, Rahman AKMF, Macassa G.

Parental socio-economic position and suicidal ideation among adolescents in

Rural Bangladesh. Journal of Psychiatry and Behavioral Sciences 2018;

4:1018.

xii

List of Abbreviations

APTA

Asia-Pacific Trade Agreement

BHIS Bangladesh Health and Injury Survey

CDC

Centers for Disease Control and Prevention

CSF Cerebrospinal

fluid

CIPRB

Centre for Injury Prevention and Research, Bangladesh

ESCAP

Economic and Social Commission for Asia and the Pacific

HPNSDP

Health, Population and Nutrition Sector Development

Program

HSC Higher Secondary school certificate

IRB

Institutional review board

LMIC

Low- and middle-income

SI Suicidal

ideation

SAARC

South Asian Association for Regional Cooperation

SUPRE-MISS

Suicide prevention-multisite intervention study on suicidal

behaviors

SSC Secondary school certificate

WHO

World Health Organization

1

1. Introduction

Adolescence (10-19 years) is a period of major change, during which childhood transforms to adulthood. This is a period of development in three main domains: physical, cognitive and social-emotional development.

During adolescence, young people move into physical maturity. Early, prepubescent changes occur when secondary sexual characteristics appear (1,2). Cognitive development is the construction of thought processes, including learning, problem solving and decision-making that allows individual to think in wider perspective and behave accordingly (3,4).

In the area of social-emotional development, three general approaches have been identified: self-concept, sense of identity and self-esteem. Self-concept is defined as the confident and consistent expression of opinions and beliefs. Self-esteem is defined as one's thoughts and feelings about one's self-concept and identity (2). Identity development involves self-recognition and self-development; obstacles to this come from family and society (5).

Adolescent development is hindered not only by external forces but by internal conflicts, including egocentrism and identity crisis (6). Most theories state that there is an overwhelming desire, across all genders and ages, to maintain, protect and enhance individual self-esteem (7).

Boys tend to be more independent and guard their self-esteem to a greater extent than girls (8). They enjoy self-esteem in supportive relationships with friends, more specifically with girlfriends. On the other hand, both girls and boys suffer low self-esteem, depression and hopelessness from broken relationships. Studies suggest that women are twice as likely to experience depression, though men are three to four times more likely to commit suicide (9,10).

Suicidal ideation, thinking about or planning suicide, is an important indicator of mental health. During adolescence, there is a risk of potential self-harming behaviors that can lead to suicidal ideation. Suicidal ideation is an expression of suffering associated with internal conflicts or depression and anxiety caused by intense physical, mental and social restructuring (3).

The definition of suicide-related behaviors is inconsistent, which has hindered progress in suicide theory and research. For example, some use the term suicidal behavior as a general term for any suicidal thought, plan, nonfatal attempt and fatal attempt. Similarly, some use the term self-harm to refer intentional self-injury without suicidal ideation (e.g. superficial skin cutting), whereas others use the term to encompass all self-injurious behaviour, regardless of intention. Because these different aspects of suicidality and self-injury can have different prevalence rates, functions, clinical

2

correlates, and outcomes, it is critical to be precise in our use of definitions and terminology (11) .

Suicidal behavior generally includes the following steps: Suicidal ideation (SI), suicide attempts and completed suicide (12). Previous studies have found a close relationship between suicidal attempts and SI (13). Suicidal ideation is a recognized precursor to suicide attempts, which may result in completed suicide (14,15). Suicidal ideation and attempts are strongly predictive of suicide deaths, but can also result in negative consequences such as injury, hospitalization, and loss of liberty, exerting a financial burden on society amounting to billions of dollars (11). However, compared to high-income countries, relatively little is known about the epidemiology of adolescent suicide and suicidal behaviors in low- and middle-income countries (16).

2. Background

2.1 Suicidal Ideation Worldwide

Suicidal behavior is a global cause of death and disability. Worldwide, suicide is the fifteenth leading cause of death, accounting for 1.4% of all deaths and an estimated 6% of all deaths among young people (16, 17). Up to 60% of adolescent and young adult suicides occur due to social factors. Youth suicide is a major global public health concern (18). In total, more than 800,000 people die by suicide each year, and it is the second leading cause of death among those aged 15 to 19 years. Low- and middle-income countries are home to more than 90% of the world’s children and youth and account for over 75% of global suicide deaths (16).

Across 17 European countries, the lifetime prevalence of ideation among students aged 15 to 16 years ranged from 15% (Armenia) to 31.5% (Hungary) (19). Across 49 low- and middle-income countries, 15.3% of adolescents aged 13 to 15 years had seriously considered suicide in the past year (20). The South-East Asia Region and Western Pacific Region had a relatively low prevalence of suicidal behaviors across genders, 10.7% for ideation and 5.0% for ideation with a plan. Major risk factors for youth suicidal behaviors include being female, exposure to bullying and violence, alcohol and drug use, mental disorders and weak family and peer relationships (16) .While much of this evidence comes from Europe and North America, recent research has expanded the knowledge of the determinants of youth suicidal behaviors in several low- and middle-income countries. Many factors associated with youth suicidal behaviors in low- and middle-income countries overlap with established risk factors from high-income countries. These include bullying, physical and sexual abuse, mental disorders and depressive symptoms, substance use and weak family and social relationships (21,22).

3

In the United States, suicide has become the second leading cause of death in adolescents (23). In Brazil, adolescents from public and private schools in the city of Recife were assessed, and 34.3% were found to have suicidal ideation or have attempted suicide. Suicide attempts had a significant association with depressive symptoms (24). Depression is one of the leading risk indicators for suicide. Over the course of the lifespan, people with untreated depression have a nearly 20% risk of suicide; about two-thirds (2/3) of people who commit suicide are believed to be depressed at the time of their deaths (25).

In this context, identifying individuals engaged in SI is critical to preventing youth suicide. Identifying factors that predict suicidal thoughts in adolescents may thus help predict subsequent suicide attempts and prevent youth suicides. Previous research indicates that factors that predict SI are multi-faceted and more clearly noticeable among adolescents than adults (26).

Primary predictors include personal characteristics, such as gender and age. According to an Australian study, more female adolescents engage in SI than male adolescents, but more male adolescents attempted suicide (27). Nonetheless, other research has not found gender to be a predictor of SI or suicide attempts. Another study found that SI increased gradually until mid-adolescence then decreased slightly. Adolescents were 12 times more likely to attempt suicide if they engaged in SI at the age of 15, with no gender differences (28). Musci et al. reported that suicidal ideation peaked in the seventh grade with a steady decline in ideation in subsequent grades (13). It remains uncertain if risks peak at different ages depending on gender, though SI and suicide attempts and completions generally rise during later adolescence (29). The most notable difference is that suicide rates are not distributed evenly across groups of people or places. For example, high-income countries have higher suicide rates than low- and middle-income countries (LMICs) at 12.7 versus 11.2 per 100,000, respectively. LMICs, however, account for over 75% of all suicides worldwide (11). 2.2. Suicidal Ideation in Developing Countries and Asia

Suicide rates have increased globally over the past 45 years among young people (10-24 years) (30). Suicide rates are higher in South Asia (31). Within Sri Lanka, the rate is exceptionally high, at more than 35 per 100,000 residents (32, 33). Nepal also has a high rate, at 25 per 100,000 residents, and in terms of gender, suicide rates among men and women are 30 and 20 per 100,000 residents, respectively. These are estimates from locally based research, which greatly exceed the ‘official’ WHO rates. Furthermore, it has been observed that unmarried women aged 10 to 24 years are at the highest risk of suicide in Nepal (34) .

Despite recent decreases, child and adolescent suicide in Japan remains a serious social problem: over 600 cases are reported per year (35). In South Korea, adolescent

4

suicide is the most critical persistent psychological health problem. Since 2001, suicide has never dropped below the second-leading cause of death among girls and boys aged 10 to19 years, with a rate of 5.3 persons per 100,000 in 2001, rising to 7.2 persons per 100,000 in 2014 (36). In South Korea, 17% of adolescents who commit suicide have very low socioeconomic status (37). Research indicates that individual psychological characteristics and depression are the most powerful predictors of suicidal ideation (38). Previous research also suggests a positive relationship between daily stress levels and SI in adolescents (39). Adolescents in South Korea typically encounter highly competitive educational environments and college entrance examinations compared to other countries, which significantly affect rates of SI (40). Further, some research suggests that sleep duration significantly predicts SI even after controlling for depression, drug use and other demographic factors, and mediates the relationship between psychological burden and suicidal risk (12). Finally, recent research has suggested a relationship between SI and behavioral problems (e.g. alcohol drinking, smoking and sexual activity) among adolescents. Adolescents aged 13 to16 years who had used substances before early adolescence were more likely to engage in SI; such problem behaviors are likely to be associated with depression and may stimulate thoughts of suicide (41).

Studies show that female adolescents experience more depressive symptoms, earlier pubertal onset and more advanced psychological maturity than male adolescents (42, 43). One study showed that girls were more susceptible to suicidal ideation than boys, particularly girls in junior school (44). Another study revealed that participants were less likely to report SI after entering high school; this may reflect the fact that students tend to establish their identity and find greater psychological stability at this age. Home environment, socioeconomic status, and living with one's parents affected the likelihood of SI. These findings indicate that adolescents with low socioeconomic status are more likely to experience hardship, potentially presenting a distraction from schoolwork and leading to an uncertain future. Private education costs are high in South Korea; therefore, students with low socioeconomic status may not feel as supported in their education, exacerbating the relationship between depression and SI. Family instability also predicts reduced support for adolescents, and it has been shown that social and family support is a strong predictor of SI in teenagers (35).

Adolescents growing up in rural areas are often conflicted when trying to balance between their attachment to their rural lifestyle, community support and their desire to gain educational, social, and occupational experiences available exclusively in urban areas. For adolescents who experience serious issues while adjusting to the schooling system, lack of attachment to the community can be a compounding risk factor (45).

5

2.3. Suicidal Ideation in Bangladesh

Bangladesh is a country in Southern Asia and is located on the northern side of the Bay of Bengal, sharing a border with India on all sides and a small border with Burma. It is a member of the Asia-Pacific Trade Agreement (APTA) and the South Asian Association for Regional Cooperation (SAARC). The current population of the country is 164,669,751 as of 2017 (46), and it has an area of 147,570 square kilometers (47). In 2019, the GDP per capita was USD 1,856. Between 2000 and 2018, Bangladesh experienced a rise in annual GDP growth per capita from 3.3% to 7% (48).

Nearly 70 percent of the population and most of the underprivileged live in rural areas of Bangladesh. Agriculture is the main occupation, although it generates less combined income than non-agricultural sectors (49). The literacy rate is around 69% (50). Access to electricity in rural areas is very poor, and rural residents routinely experience supply disruptions (51). There is lack of access to water and quality sanitation, which results in the prevalence of various diseases (e.g. diarrhea) (52). Rural areas also experience higher levels of poverty compared to urban areas (53).

Patriarchal families, early marriage and dowries are common in rural Bangladesh (54). Reasons behind early marriage are a lack of social security for women, lack of awareness, attempt to avoid paying a dowry, negative impact of giving scope for marriage by likings, lack of application of marriage laws, scarcity of expected bridegroom, protecting family prestige and status (55). _{Child abuse is also common in}

rural Bangladesh (56). In recent years, rural Bangladesh has seen a growing economy in the agricultural sector, non-agricultural sector and through remittance, which has enabled better access to quality education. However, maintaining the sustainability of these opportunities and quality education are still great challenges (57). Despite these recent developments, rural residents still retain animistic beliefs.

Bangladesh has an adolescent (aged 10 to 19 years) population of approximately 36 million, which is more than one-fifth of the total population of the country. The main challenges experienced by the adolescent population are poverty, lack of access to information and services, negative social norms, inadequate education, social discrimination, child marriage and early child-bearing (58). The rate of child marriage in Bangladesh is 59%, and it ranks first in Asia and fourth in the world in the prevalence of child marriage (59). Among rural adolescents, the literacy rate is around 61%. In addition to their studies, rural adolescents do household work. They also have limited exposure to media and very little leisure time. The government and NGOs provided various forms of income generating skills training for rural youth, and they are breaking into the labor market in the rural economy (60,61).

6

Fig 1. Picture of rural Bangladesh

Studies have shown increasing rates of violence among unmarried adolescents in rural Bangladesh (62,63). Gender, socioeconomic status and location (districts) are the factors that influence adolescent violence in selected rural districts (64). Adolescents and youth in rural Bangladesh are particularly vulnerable to health risks like malnutrition, infection, reproductive health problems, psychological problems and drug addiction. (65-67). This is largely due to their lack of access to information and services as well as societal pressure to take on the same role as adults notwithstanding the physical, mental, and emotional changes they are undergoing. Suicide appears to have increased in Bangladesh, especially among rural adolescents (8). In 2005, the nationwide data of the Bangladesh Health and Injury Survey (BHIS) on Children found that suicide was the leading cause of death for older adolescents of both sexes, with the highest rate of 50 deaths per 100,000 among 17-year-old girls (68). Suicide was the fifth most common cause of death overall and the leading cause of death among adolescents. In 1998, the Economic and Social Commission for Asia and the Pacific (ESCAP) found that nearly 30 of 100,000 young adults in rural Bangladesh committed suicide every year (8). The Bangladesh Health and Injury Survey reported that more than 2,200 children committed suicide in the year 2003 – or about six per day. Of those six, four were female. Suicide is the number one cause of death among this age group (68).

7

Fig 2. Adolescents of rural Bangladesh

A systematic review of 23 studies on suicide and suicidal behaviors in Bangladesh has shown that the suicide rate is higher among younger women. The rate in adolescent girls is exceptionally high by international standards, and reflects the effects of poverty, low status, violence directed towards girls and women, and the forced marriage of young, teenage girls (69).

3. Determinants of Suicidal Ideation

among Adolescents

The determinants of suicidal ideation can be classified as biological, psychosocial, environmental and socio-economic.

3.1 Biological and Psychological Determinants

Biological Risk Factors

The limbic system: It is the major primordial brain network, which regulates biological functions in line with our mood/emotion. It is a network of regions that work together to process and make sense of the world (70).

8

The physiological stress system: The hypothalamic-pituitary-adrenal (HPA) axis is one of the body’s major systems for modulating physiological responses to actual, anticipated, or perceived harm and plays a major role in adaptation to stresses of all types.

Neurochemistry: The monoamines, particularly dopamine, norepinephrine and serotonin, have been the focus of much of the research on mental disorders. These neurochemicals show significant changes in various neuropsychiatric disorders. Cerebrospinal fluid (CSF) Levels: Metabolite (5-HIAA) levels in CSF are a strong correlate of current and future suicidal behavior.

Genetic Factors: Similar to most complex conditions, such as obesity, hypertension and coronary artery disease, there is growing evidence that genetic factors are related to susceptibility for suicidal behavior (71).

These factors are associated to a higher degree with female adolescents, who report higher rates of depression, anxiety, suicidal thoughts, ideations, and suicide attempts; however, male adolescents are four times more likely to die by suicide. One explanation for this contradiction is that males tend to use more lethal means to commit suicide (such as firearms), whereas females choose methods that are more responsive to medical interventions (such as drug overdose or poisoning) (72).

There is an increased propensity for risk-taking in adolescence, and they are more likely to engage in risky behaviors. During adolescence, there is an increased interest in peer relationships (73). Susceptibility to peer influence increases during the early teen years and peaks at about the age 14 (74). Consistent with observable changes in peer relationships, brain-imaging studies have shown that several areas of the brain make adolescents more sensitive to the rewards of peer relationships than adults (75). This causes teens to focus on their peers in decision-making situations that involve risky behavior. Research shows that adults use this brain region more heavily at times of social exclusion than adolescents (76). During adolescent years, however, this brain region is still developing (77), so adolescents may not be able to control distress as effectively during times of peer social exclusion. This likely contributes to risky behaviors to cope with peer exclusion. Adolescents who engage in multiple risk behaviors show a higher risk for suicide (78).

Psychological Risk Factors

Mental health problems and psychiatric disorders are known to be risk factors for suicidal behavior among adolescents (72). Psychologists agree that adolescence is inherently a time of disturbance and psychological confusion (2). Depression is the most common disorder associated with adolescent suicide and suicidal behaviors and often overlaps with other diagnosed disorders. For example, prior suicide attempts among adolescents increase the chances of subsequent suicidal behaviors and death.

9

Research also shows that adolescents who engage in suicidal behaviors tend to behave more aggressively and impulsively, lack positive coping and problem-solving skills, feel less supported by family and friends and have a general feeling of hopelessness. Although many of these behaviors can be associated with psychiatric disorders, it may be important to assess these behaviors independent of a psychiatric diagnosis.

Substance abuse also affects rates of adolescent suicide. The combination of mental health problems and substance use may be even more dangerous. Particularly among males, the existence of conduct disorder, substance abuse and depression can be fatal. Limiting access to drugs and alcohol can prevent adolescent suicide (72).

3.2. Environmental Determinants

Among adolescents, a stressful life event often precedes a suicide attempt (79). Although not the cause of the suicide attempt, these events may make an adolescent feel more vulnerable, contribute to a feeling of hopelessness and despair, or cause the adolescent to become overwhelmed or act impulsively. Relationship breakups, parental divorce, death of a loved one, military deployment of a parent, academic failure, and physical/sexual child abuse are events known to precede a suicide attempt (80,81).

Childhood sexual abuse (CSA) is one of the most eminent risk factors for suicidal behaviors all over the world. About 20% of women and 8% of men report a history of CSA (82).Growing global connectivity has also made it possible for new global crimes to emerge, and one of the most devastating is online child sexual abuse and exploitation, including child pornography and commercial sexual exploitation; enticement, solicitation, and grooming; bullying, harassment, and cyber-stalking. Online child sexual abuse is currently on the rise in Bangladesh (82–85). Although most adolescents experience these situations at some point, research suggests that in combination with a psychiatric disorder, many of these negative life stressors can greatly contribute to suicide risk and behaviors (86).

Still, several environmental factors can serve to protect young people. For instance, positive parental relationships are one of the most consistent protective factors. Adolescents who are more connected and supported by their family have a lower risk of engaging in suicidal behaviors (78).

Peer relationships and school environment also seem to play a role in adolescent suicide. During adolescence, the primary attachment shifts from parental figures to peers. These relationships can influence adolescents in a variety of ways. Having poor social skills, low self-concept and social self-concept, and feeling rejected or isolated by peers can be risk factors for adolescents (87).

10

Finally, recent studies have started investigating the impact of technology on adolescent suicide risk. It has long been recognized that media coverage on suicide can lead to suicide clusters (an excess number of suicides or attempts than would be expected in a particular community at a particular time). For young people between the ages of 15 and 24, it is estimated that up to 13% of suicides occur in clusters (88). Further, the internet can encourage suicide and provide youth with information about how to commit suicide successfully, though it also allows youth to more easily access information about where and how to get help (89).

3.3. Socio-Economic Determinants

The health of adolescents is strongly affected by social factors at personal, family, community and national levels. Nations present young people with structure in the form of opportunities as they mature. The strongest determinants of adolescent health worldwide are structural factors, such as national wealth, income inequality and access to education. Furthermore, safe and supportive families and schools, together with positive and supportive peers, are crucial to helping young people develop their full potential and attain the best health in the transition to adulthood. Improving adolescent health requires improving daily life with families and peers in schools as well as addressing risk and protective factors in the social environment (90).

Adolescents with low socioeconomic status or who live with one parent were more likely to engage in SI. Interestingly, these populations have shown higher SI levels than adolescents living with foster parents (35).

Some studies have examined suicide rates according to socioeconomic differences. Variables identified were educational level, socioeconomic status and income level. Feroz et al. found that 45.7% of those who completed or attempted suicide were from lower-income households, compared to 37.1% in the lower middle class and the fewest (17.2%) in the middle and upper-class. According to the same study, families with low incomes (less than 5000 Taka, or less than about $60 a month) had the highest rates of suicide. Socioeconomic status (measured by a composite score of earning capacity, housing status and possession of essential and luxury goods in the family) marked a clear socio-economic gradient in completed and attempted suicide cases: 68.4% of such cases occurred in the lower classes, compared with 7.3% in the ‘upper class’ (91). Reza et al. observed that 80.5% of all suicide cases were from lower (<3000 Taka, i.e. about $30 per month) or lower-middle (3000–5000 Taka per month) income groups (92). In a national survey in 2003, Mashreky et al. observed that the majority of suicide victims (55.0%) were found to be economically very poor, with a monthly family income of less than $50 a month, and 14% of the suicide victims’ family members earned less than $25 a month. The same study also found that illiteracy is a strong correlate of suicide (42% of the suicide victims were illiterate) (68).

11

When children go through adolescence, there is often a significant increase in parent– child conflict and a less cohesive familial bond. Arguments often concern minor issues of control, such as acceptable clothing, and the adolescent's right to privacy (92), which adolescents may have previously viewed as issues over which their parents had complete authority (30). Parent-adolescent disagreement also increases as peers have a greater impact on one another, introducing new influences that may be in opposition to parental values. Adolescents who have a good relationship with their parents are less likely to engage in various risk behaviors, such as smoking, drinking, fighting and/or unprotected sexual intercourse (93).

4. Conceptual Model

The present thesis uses the socio-ecological model (SEM) to identify the social determinants of adolescent suicidal ideation. Socio-ecological models are used in various fields, such as the ecological model for health promotion and the family process conceptual model for identifying risk factors of diseases and suicide prevention (94). The SEM is a four-tier model used to organize risk and protective factors that facilitate corresponding strategies. The model builds from macro to micro levels within a four strata tier: societal, community, relational and individual levels. The “societal” factors are concerned with the larger scale issues, such as social and cultural norms, policies and other guiding rules or laws. The “community” level influences can be described as those factors associated with neighborhood support or organizations (e.g. schools, workplaces and healthcare providers). Furthermore, “relational factors” are those defined by person-to-person interaction, such as social support or withdrawal, peers and family. Last, the “individual” level factors are related to personal characteristics, such as demographics, attitudes and health conditions, among others (95,96). The socio-ecological model is considered to be important for a multilevel public health approach to suicide prevention (33, 97). In this thesis, the socio-ecological model (its relational level) is used specifically to understand how parental socioeconomic status (relational level of SEM) measured through education, occupation, marital status, income and home ownership, impacts adolescent suicidal ideation in the context of rural Bangladesh.

5. Rationale

Adolescence is a period in life when individuals have specific health and developmental needs and rights. It is also a time to develop knowledge and skills, learn to manage emotions and relationships, and acquire attributes and abilities that will be important for enjoying the adolescent years and successfully transitioning to adulthood (97, 98).

12

More than half of all adolescents globally live in Asia. In absolute numbers, South Asia is home to more adolescents (around 340 million) than any other region. It is followed by East Asia and the Pacific, with around 277 million. The adolescent population of either of these regions dwarfs that of any other region in the world (99). The latest Bangladesh National Census in 2011 estimated that there are approximately 36 million adolescents in the country, constituting one fifth of the total population of 150 million. Globally, millions of adolescents die or become sick from preventable causes. An estimated 1.2 million adolescents die each year, over 3,000 every day. Half of all mental health disorders in adulthood start by age 14, but most cases are undetected and untreated (100,101).

The Lancet Adolescent Health series reported that, ‘Failure to invest in the health of the largest generation of adolescents in the world’s history jeopardizes earlier investment in maternal and child health, erodes future quality and length of life, and escalates suffering, inequality, and social instability’. The lack of a gender and needs responsive adolescent health framework, inadequate investments in the healthcare sector, or the lack of appropriate engagement and participation of adolescents themselves as rights holders will lead to critical gaps in a country’s overall development trajectory (18).

Adolescent health (AH) is one of the major items on the agenda for the government of Bangladesh. Adolescent health has been incorporated into the Adolescent and School Health Program of the Directorate General of Health Services (DGHS) in the Health, Population and Nutrition Sector Development Program (HPNSP), 2017-2022 (102). Adolescent suicide is a major public health problem. Suicide appears to be increasing in Bangladesh, especially among rural adolescents (103). However, the prevalence and factors for suicidal ideation are not well researched in low- and middle-income countries, especially in Bangladesh (103,104). The generation of evidence on magnitude and risk factors will help to design interventions for the prevention of adolescent suicide in Bangladesh as well as in the LMIC context.

There is a lack of population-based studies addressing suicidal ideation among adolescents. The aim of the present thesis is to investigate the lifetime prevalence and social determinants of suicidal ideation among adolescents aged 14 to 19 years in rural Bangladesh. The findings of this thesis will serve as a baseline for a nationwide study in order to develop a suicide prevention program for adolescents and young adults. The present study will facilitate the development of an appropriate model for adolescent suicide prevention and help policy makers at various levels in their efforts to decrease adolescent suicide in Bangladesh as well as other countries with a similar socio-cultural and geographical context. Finally, the data collected will be utilized for estimations and comparisons with findings from other countries and further research in the field.

13

6. Objectives

6.1 General Objective:

The main objective of the thesis is to assess the magnitude and social determinants of suicidal ideation among adolescents in a rural area of Bangladesh.

6.2. Specific Objectives

a) To estimate the prevalence of suicidal ideation and related factors among adolescents in rural Bangladesh.

b) To determine the role played by parental socio-economic status in suicidal ideation among adolescents in rural Bangladesh.

7. Method

7.1. Setting and Participants

A population-based cross-sectional study was conducted in the Raiganj sub districts of Sirajganj district, located in the northwest region of Bangladesh. The population of Raigaonj is estimated to be around 317,000 of which 30,242 are adolescents aged 14 to19 years. The main source of employment in Raiganj is agriculture (69.27%), with others being business, handicrafts, service and non-farm labor. The average literacy rate is about 36%. Primary healthcare centers operated by the government and public sectors are available in the region (104).

This area was chosen because the Centre for Injury Prevention and Research, Bangladesh (CIPRB) had established a field laboratory at Raiganj, through which a good relationship has been developed between the local community and CIPRB. The present study was able to take advantage of CIPRB’s presence to collect sensitive information regarding suicidal ideation among adolescent boys and girls in this area. The study was conducted in three unions out of the nine in which CIPRB has maintained an injury and demographic surveillance system since 2005 (105). The surveillance data was used to identify and select the adolescents for interview in the community. However, demographically and geographically, rural Bangladesh is almost homogeneous.

14

Fig 3. Map of area studied.

7.2. Study Sample and Design

A cross-sectional descriptive design was used for the survey. The sample size was calculated assuming the prevalence of suicidal ideation among adolescents at 0.04 and a precision of 0.008 (20% of prevalence). The calculated sample size was 2,304. Considering an approximate refusal rate of 8%, the survey recruited 2,500 adolescents and interviewed 2,476 (response rate 99%). There were a few refusals and some respondents were not available during data collection. Age and sex were used as matching factors within the same block.

7.3. Sampling Procedure

Surveillance data from the Centre for Injury Prevention and Research, Bangladesh (CIPRB) was used to develop the sampling frame. The surveillance procedure used by CIPRB has been published previously (105). Within the CIPRB surveillance area, a total of 30,242 adolescents aged 14 to 19 years were identified. A total of 2,500 adolescents were then selected from all adolescents. The surveillance area within Raiganj was divided into 19 blocks for regular data collection. Out of 19 blocks, 6 blocks were selected randomly for this study. From each selected block, 417 adolescents were randomly selected. The name and address of each selected adolescent was

15

given to a data collection team who would conduct an interview. Ultimately, a total of 2,476 adolescents were recruited to the study.

7.4. Survey Procedure

The research team visited the selected adolescents and their families and informed them about the study. Families who provided consent and respondents who agreed to take part in the study were interviewed face-to-face, using the SUPRE-MISS questionnaire (adapted according to the Bangladesh context).

The original SUPRE-MISS questionnaire is an instrument developed by the WHO and has been validated both in developed and developing countries. The overall goal of SUPRE-MISS was to aid in the reduction of morbidity and mortality associated with suicidal behaviors. More specifically, its objectives were: (a) to increase awareness of the burden of suicidal behaviors; (b) to identify reliable and valid variables for determining risk factors for fatal and non-fatal suicidal behavior, with a special emphasis on social factors; (c) to describe patterns of suicidal behavior; (d) to identify variables that determine the presentation, at healthcare facilities following a suicide attempt; (e) and to improve the efficiency of general healthcare services through the identification of specific interventions that effectively reduce suicide attempts (106). For the purposes of this thesis, the original SUPRE-MISS questionnaire was translated into the local language “Bangla” (and back-translated to English) and adapted by a group of public health professionals and psychologists working at Dhaka University. The questionnaire covered relevant background and socio-demographic information to identify the magnitude and social determinants of suicidal ideation among adolescents in a rural area of Bangladesh. It was pretested before finalization. Data were collected in two phases. In the first phase, adolescents with suicidal ideation were identified using the screening questionnaire. Two local researchers collected data from the households. They assured respondents and their caregivers about the confidentiality of the survey. Moreover, they encouraged guardians to ensure privacy during data collection so that respondents did not hesitate to speak with researchers and provide relevant information. Participation was voluntary, and confidentiality was emphasized. Each caregiver and respondent signed an informed consent. Codes were placed on the questionnaire rather than names.

In the second phase, a clinical psychologist collected detailed information from the adolescents who reported suicidal ideation followed by a counselling session at the CIPRB office.

16

7.5. Measurement of Variables

7.5.1. Dependent Variable

The outcome variable of the present study was lifetime suicidal ideation. This was assessed using the following question: “Have you ever seriously thought of committing suicide?” (Estimate lifetime prevalence). The answers to the question were dichotomized. Respondents who answered “yes” were regarded as having suicidal ideation, and those who answered “no” were regarded as not having suicidal ideation.

7.5.2. Independent Variables

The independent variables (demographic and socio-economic) were age, sex, marital status, education, occupation, monthly family income and expenditure, home ownership, wealth index; parental occupation, education and marital status; parents living or deceased, other residents in adolescent’s home.

Age was divided into three age groups for this study: 14–15, 16–17, and 18–19 years. Marital status was defined in terms of three groups: married, unmarried, divorced/separated. Education level was assessed in a range from illiterate to the highest level of formal education. Four levels of education were created: illiterate, primary school or equivalent, secondary school certificate (SSC) or equivalent, and higher secondary school certificate (HSC) or equivalent. Regarding work status, few of the adolescents are formally employed. As such, the adolescents are grouped into 4 categories: student, service sector, household work/unemployed, and day laborer. Family income was measured as income of all family members as reported by the adolescents. Three groups were created for total monthly family income: up to BDT 10,000 (Bangladeshi Taka), 10,001–20,000 and 20,001and above. The Wealth Index was grouped into five statuses: lowest, second, middle, fourth and highest.

Four levels were created to measure parental education: illiterate, primary school or equivalent, secondary school certificate (SSC) or equivalent and higher secondary school certificate (HSC) or equivalent. Occupation was grouped into four categories: service sector, business, and agriculture/housework and day laborer. Parent’s marital status was defined as married or single parent.

7.6. Statistical Analyses

The demographic characteristics were calculated using descriptive statistics, such as frequencies and percentages. Comparisons of proportions between groups were carried out using χ2 tests. Furthermore, multivariate logistic regression was used to identify the relation between various co-variants and lifetime occurrence of suicidal thoughts among adolescents (paper 1). Additionally, descriptive statistics and a logistic regression analysis were used to examine the associations between parental

socio-17

economic variables and suicidal thoughts in adolescents (paper 2). When adjusting for potential confounders, results were presented as odds ratios (OR) with 95% confidence intervals. The significance level was set at < 0.05. All analyses were performed using SPSS 20.

7.7. Ethical Considerations

This research was approved by the Institutional Review Board (IRB) of the Center for Injury Prevention and Research, Bangladesh (CIPRB).

18

8. Results

8.1. Study 1

The lifetime prevalence of suicidal ideation (SI) was 5% among adolescents, with a majority being female 66 (52.8%), unmarried 103 (82.4%) and students 92 (73%).

Table 1: Socio-demographic characteristics of the respondents (n= 2,476)

Variables Total population, N (%) Mean ± SD

Age years (mean±sd) 16.2±1.4

Age groups (years)

14- 15 919 (37.1%) 16- 17 1,005 (40.6%) 18-19 552 (22.3%) Sex Male 1,209 (48.8%) Female 1,267 (51.2%)

Education years (mean±sd) 6.7±2.8

Education status Illiterate 79 (3.2%) Up to primary 932 (37.6%) Up to SSC 1,165 (47.1%) Up to HSC 300 (12.1%) Occupation Unemployment 51 (2.1%) Student 1,383 (55.9%) Farming 65 (2.6%) Weaver 182 (7.4%) Household work 493 (19.9%) Day labourer 124 (5%) Rickshaw puller 20 (0.8%) Service 111 (4.5%) Business 46 (1.9%) Marital status Unmarried 1,944 (78.5%) Married 530 (21.4%) Divorced/Separated 2 (0.1%)

Monthly family income in Taka

(mean±sd) 10,194±4,743

Monthly household Family Income

Up to 10,000 tk 1,887 (76.2%)

10,001 to 20,000 tk 546 (22.1%)

19

Suicidal ideation was significantly associated with age. Adolescents aged 18 to19 years had odds ratios of 2.31 with a CI of 1.46–3.65 after adjusting for the co-variates education, occupation, living with whom and home ownership. There is a significant association between education and SI where confounders were age, occupation, living

Table 2: Prevalence of suicidal ideation according to socio-demographic variables

Socio-Demographic variables Ideation n (%) Confidence Interval

Life time prevalence 125(5%) 4.21-5.98

Age (mean±SD) 17±1.3 yrs Age group 14- 15 yrs 34(3.7%) 2.58-5.13 16- 17 yrs 46(4.6%) 3.37-6.05 18-19 yrs 45(8.2%) 6.00-10.75 Gender: Male 58(4.8%) 3.66-6.15 Female 67(5.3%) 4.12-6.66

Mean year of education: 8.28±2.5 years Education status Illiterate 1(1.3%) 0.032-6.85 Up to primary 25(2.7%) 1.74-3.93 Up to SSC 69(5.9%) 4.63-7.43 Up to HSC 30(10%) 6.84-13.96 Occupation: Student 86(6.2%) 5.00-7.62 Service 00 00 Household work/unemployed 32(5.9%) 4.05-8.20 Day labourer 7(1.8%) 0.72-3.65 Marital status Unmarried Married Home ownership No Yes 106(5.5%) 18(3.4%) 6(27.3%) 118(4.8%) 4.48-6.55 2.02-5.31 10.73-50.22 4.01-5.59

Monthly household Family Income category:

Up to 10000 tk 104(5.5%) 4.53-6.63

10001 to 20000 tk 20(3.7%) 2.25-5.60

20001 and above 1(2.5%) 0.06-13.16

Monthly family expenditure category:

Up to 10000 tk 110(5.4%) 4.44-6.45

20

with whom and house ownership. Adolescents with secondary school certificate had odds ratio of 2.38 with CI 1.51–3.77 and higher secondary school certificate or higher education had odds ratios of 4.15 with CI 2.41–7.1. SI was significantly associated with occupation after adjusting for confounders such as age, living with whom and home ownership (day laborers had an OR of 0.28 with a CI of 0.13–0.60). Living without parents is a predictor of SI in adolescents (OR of 1.80 with a CI of 1.07–3.03) even after controlling for age, education, occupation and home ownership. There is a significant association between home ownership (owning a home had an OR of 0.14 with a CI of 0.05–0.35) and SI after adjusting for confounders such as age, education, occupation and living with whom.

8.2. Study 2

Most of the parents had only a primary school education (mothers 58.7% and fathers 49.5%). The majority of them were farmers (53.3% of fathers) and homemakers (96.5% of mothers). Monthly income and household expenditure amounted to BDT 10,000 or

Table 3: Suicidal ideation and socio-demographic characteristics of adolescents

Variables N (%) Unadjusted Adjusted

OR 95%CI OR 95%CI Age 14-15 yrs 34(3.7%) Reference 16-17 yrs 46(4.6%) 1.25 0.79-1.96 1.04 0.67-1.62 18-19 yrs 45(8.2%) 2.31 1.46-3.65 1.80 1.10-2.94 Education

Primary and lower 26(2.6%) Reference

Secondary 69(5.9%) 2.38 1.51-3.77 1.68 1.06-2.68 Higher secondary and above 30(10%) 4.15 2.41-7.14 2.22 1.21-4.09 Occupation Student 86(6.2%) Reference Household work 32(5.9%) 0.94 0.62-1.43 1.02 0.63- 1.65 Day laborer 7(1.8%) 0.28 0.13-0.60 0.32 0.13- 0.77 Home ownership No 6(27.3%) Reference Yes 118(4.8%) 0.14 0.05- 0.35 0.12 0.04-0 .34

Living with whom

Parents 106(4.7%) Reference

21

118 USD. Suicidal ideation was more common among adolescents of low-income parents (104, 5.5%) and adolescents who did not live with their parents (18, 8.2%).

Table 1: Socio-economic attributes of the adolescent’s parents

Socio-economic variables N (2476 ) Percentage (%) Mother’s Education Illiterate 72 29.00% Primary 1424 58.70% Secondary 248 10.20% HSC and above 31 1.30% Father’s education Illiterate 752 31.90% Primary 1169 49.50% Secondary 326 13.80% HSC and above 113 4.80% Mothers Occupation Service 31 1.30% Business 37 1.50% Agriculture/House work 2356 96.50% Day laborer 17 0.70% Fathers Occupation Service 179 7.60 % Business 523 22.10% Agriculture/House work 1262 53.30%

Day laborer/Rickshaw or van puller 403 17%

Parents Marital Status

Married 2321 93.70% Single Parent 155 6.30% Living With Parents 2249 91.10% Others 220 8.90% House Ownership No 22 0.90% Yes 2,454 99.10%

Monthly Household’s family income

Upto 10000tk 1887 76.20%

10001-20000tk 546 22.10%

20001- above 43 1.70%

Monthly Family Expenditures

Upto 10000tk 2044 82.60%

22

Wealth Index Status

Lowest 505 20.40%

Second 527 21.30%

Middle 454 18.30%

Fourth 524 21.20%

Highest 466 18.80%

SI was found to be significantly associated with parental education, marital status and home ownership. Not being able to live with their parents was also a significant factor. Adolescents of parents reaching the SSC level of education had an odds ratio of 2.10 (1.21, 3.64) and 1.92 (1.15, 3.23) for mothers and fathers respectively. Neither parental income nor household expenditure was associated with adolescent suicidal ideation. Suicidal ideation among adolescents with single parents had higher odds (OR 3.00, CI 1.75-5.19) in comparison to adolescents who had both parents. Adolescents whose parents owned a home and adolescents who did not live with their parents had odds ratios of 0.14 (0.05, 0.35), and 1.80 (1.07, 3.03) respectively. After adjusting for other covariates, parental marital status and home ownership were found to be significantly associated with adolescent suicidal ideation.

Table 2: Prevalence of Suicidal ideation according to parent’s SES

Characteristics Suicidal Ideation (N= 125) Prevalence (%)

Parents Marital Status (n=124)

Married 107 4.60% Single 17 10.96% Fathers Occupation (n=122) Service 9 5.00% Business 25 4.80% Agriculture/ Housework 61 4.80%

Other (Day Laborer/ rickshaw or van puller

27 6.70%

Mothers Occupation (n=122)

Service & Business 4 5.90%

Agriculture/ Housework 115 4.90%

Other (Day Laborer/ rickshaw or van puller

3 17.70%

Education of Mother (n=122)

Illiterate 34 4.70%

Primary 64 4.50%

23

Education of Father (n=123) Illiterate 34 4.50% Primary 57 4.90% Secondary 28 8.60% HSC and Above 4 3.50% Income (n=124) Up to 10000 104 5.50% 10001-20000 20 3.70% Expenditure (n=125) Up to 10000 110 5.40% 10001-20000 15 3.70%

Have your own house (n=124)

No 6 27.30%

Yes 118 4.80%

Living with whom (n=124)

Living With Parents 106 4.70%

Living With Others 18 8.20%

Table 3: Association between parent’s socio-economic status and suicidal ideation among adolescents Unadjusted Adjusted Variables Suicidal Ideation, n (%) OR 95 % CI OR 95 % CI Mother’s education Illiterate 34(4.7%) Reference Primary 64(4.5%) 0.97 0.63, 1.48 0.87 0.50, 1.50

Secondary and above 24(8.6%) 2.10 1.21, 3.64 1.69 0.87, 4.12

Father’s education

Illiterate 34(4.5%) Reference

Primary 57(4.9%) 1.08 0.70, 1.66 1.16 0.66, 2.01

Secondary 28(8.6%) 1.92 1.15, 3.23 1.39 0.71, 2.72

HSC and above 04(3.5%) 0.82 0.29, 2.36 0.50 0.15, 1.733

Parent’s Marital status

Married 107(4.6%) Reference

Single 17(11.0%) 3.00 1.75, 5.19 2.35 1.31, 4.20

Living with whom

Living with parents 106(4.7%) Reference

Living with others 18(8.2%) 1.80 1.07, 3.03 1.51 0.87, 2.65

Having own house

No 6(27.3%) Reference

24

9. Discussion

9.1 Prevalence of Suicidal Ideation among Adolescents and Young

Adults in Rural Bangladesh

This study found that age, education, occupation, family home ownership and living with parents were associated with suicidal ideation among rural adolescents in Bangladesh. In addition, the study found that 5% of the adolescents exhibited lifetime SI. This finding concerning the prevalence of lifetime SI is consistent with a number of previous surveys (107,109).

In an Iranian community survey, the lifetime prevalence(23%) of SI was found to be about 5 fold among adolescent compared to the present study (109); the distinction may be due to fact that the study was done in an urban setting across various age groups (15–55 years). Another study done by Nock et al. showed that the cross-national prevalence of seventeen countries for lifetime SI was 9.2 %, which is far higher than the prevalence found in the present study, which may be due to variance in samples and settings (104). Furthermore, a Southeast Asian study also found high prevalence (11.7%) of SI compared to the present study, which could be due to differences in the study sample (university students) (110). Another study carried out across 32 low- and middle-income countries showed about a 3-fold higher prevalence (16.2%), which may be attributable to the fact that the study was done in a school setting (16).

However, the prevalence of lifetime suicidal ideation was low compared to the findings of studies of adolescents in India (21.7%) and Pakistan (31.4%) (111, 112). This dissimilarity could be related the approach of the current study, which was completed in a rural community, while the Indian and Pakistani studies were done on school students in urban settings. The prevalence of SI in the current study is similar to that of Taiwan (5.3%) and higher than that of Beirut (2.1%) (113).

A higher rate of suicidal ideation was found in females. A number of previous studies have shown equivalent results (113–115). Adolescents between 18 and 19 years of age had a significantly high risk of lifetime suicidal ideation. Other studies have found that the first onset of psychiatric disorders amplified the danger of suicidal ideation at age 15 (116), and suicide rates escalate significantly among adolescents with age (117).

The current study also showed that higher education was strongly associated with suicidal ideation among adolescents in rural Bangladesh. This might be due to expectations that do not match the reality they encounter in the real world. As a result, they become frustrated, depressed and think about taking their lives. A study in Italy found that young people (15– 24) with higher levels of education were more likely to

25

consider suicide. The authors interpretation is that highly educated people may consider suicide when they encounter failure, public shame or high premorbidity (118). On the other hand, a number of studies have found contrary outcomes, such as the South Korean study by Kim et al., which shows that low education was linked with suicidal behaviors (119). A study in the United States also demonstrated that higher education and income were connected with reduced suicide rates due to improvements in intelligence and socio-economic status (120).

Students are more susceptible to suicidal ideation than household workers and day laborers. This might be due to academic pressure, relationships with peers or teachers, parental expectations for performance, etc. Adolescents who were jobless and living with family were more prone to suicidal ideation. A likely explanation is that adolescents with no formal occupation could be more susceptible to the impact of unpleasant life events. Teenagers without a job are also perceived to be a group with a high risk of suicidal behavior across many societies (121).

9.2. Parental Socio-Economic Status and Suicidal Ideation among

Adolescents in Rural Bangladesh

The study found a relationship between adolescent suicidal ideation and parental socio-economic status in rural Bangladesh. Also, parental factors such as education, occupation, marital status, adolescent living with parents or others and home ownership were associated with adolescent suicidal ideation.

Adolescents of parents having an education up to SSC had higher odds of suicidal ideation compared to adolescents with illiterate and highly educated parents. In this study, only 3%-4% of adolescents reported that at least one of their parents held a college diploma or higher, which is very low compared to another study where 54.2% of students reported that one of their parents held a college diploma or higher (122). This is because fewer people pursue higher education in Bangladesh. Other studies have shown that parental education and family income are significant predictors of adolescent suicidal ideation (123,124). According to the current study, adolescent suicidal ideation is significantly associated with parental education up to the SSC level. This could be due to the fact that expectations are high among less educated parents but they are deficient in parenting, more critical and less able to deliver proper care. On the other hand, educated parents who are more caring would certainly have a better relationship with their children, which could have a positive impact on the adolescents’ psychological well-being and lower their suicidal ideation. Illiterate parents are not as concerned with an adolescent’s academic performance, so teens of these parents have a greater degree of self-determination, which is one of the factors that promotes psychological well-being.

26

The current study observed a higher prevalence of suicidal ideation among adolescents in low-income households, but the association was not significant. Parental marital status and home ownership were significantly associated with adolescent suicidal ideation after adjusting for other covariates. Other studies suggest that parental socioeconomic status (SES) is a key factor in determining adolescent health as it has an impact on the availability of resources to maintain healthy lifestyle (125). These results are in line with those who argue that the relational level of the SEM that relates to the family context has an impact on overall adolescent wellbeing (97).

About 91% of the adolescents lived with both parents in the present study while another study reported that 87% of respondents lived with both parents, which reflects the closeness of findings. Adolescents living with parents have a protective factor, such as parental attachment, which is very important for the adolescent’s healthy behaviour. For instance, a recent study found that living apart from parents was a risk factor for suicide, even after adjusting for other risk factors (126). Studies have also observed that a lack of perceived parental support or availability was associated with suicide attempts among adolescents (127).

Family home ownership was a protective factor against suicidal ideation among adolescents and was attributable to increasing the insecurity feeling to homelessness. Similar findings were reported in a study where homeless youth were found to have higher prevalence of suicidal ideation (128).

9.3. Strengths and Limitations

The study in this thesis is the first assessment of suicidal ideation among adolescents in rural Bangladesh, which can be considered a strength of this research. In addition, the participants were selected through random sampling within a surveillance area. Also, highly trained clinical psychologists and sociologists performed the data collection. However, the thesis is not without limitations.

The present study does not focus on determinants of gender differences in suicidal ideation and trajectories of suicidal ideation, which represent essential knowledge for the prevention of adolescent suicide. Also, the findings cannot be generalized for the country as a whole even though the rural communities throughout the country can be considered homogenous. Furthermore, the studies in this thesis used a cross-sectional design, making it difficult to determine the causality of the observed associations.

9.4. Future Research

Based on the findings of this thesis, future studies to be carried out within the Bangladesh context should focus on psychosocial correlates and the trajectories of adolescent suicidal behavior. In the long run, this could provide in-depth insights that

27

can be used to design adolescent suicide prevention programs in Bangladesh and other developing countries.

10. Conclusions

The thesis found that the prevalence of lifetime suicidal ideation among adolescents is quite high in a rural community of Bangladesh. Moreover, the thesis found that the individual characteristics of adolescents and their parental marital status, education and home ownership were statistically significantly associated with suicidal ideation even after adjusting for the potential confounders.