1 Introduction

Suicide is a major global concern, ranking among the top twenty leading causes of death worldwide [1, 2]. Shockingly, it causes more deaths than malaria, breast cancer, or war and homicide combined [3]. The World Health Organization (WHO) has identified the reduction of suicide mortality as a top priority. Suicide reduction is included as an indicator in the United Nations Sustainable Development Goals (SDGs) and the WHO Mental Health Action Plan 2013–2030 [4]. Annually, suicide accounts for about one million deaths globally [5], which means 16 people per 100,000 people die by suicide, or one person dies every 40 s [2]. Suicide is the second leading cause of premature mortality in people aged 15 to 29 years worldwide [6], after traffic injuries. It is also the third leading cause of death for those aged between 15 and 44 years [7].

Suicide is defined by the World Health Organization (WHO) as an act of deliberately killing oneself with the full knowledge or expectation of its fatal outcome [1]. Survivors of suicide attempts may sustain physical injuries that require medical attention. Some of these injuries may be short-lived or permanently disfigure the affected individuals [8], and the families of the survivors often suffer from psychosocial complications [9].

Remarkably, 79% of all global suicides occur in low- and middle-income countries (LMICs) [3]. Suicidal behaviour among adolescents is higher in low- and upper-middle-income countries [10], such as 27.9% in Kenya and 21.9% in Benin Republic, compared to 17.5% in Seychelles, 10.2% in Costa Rica, and 15% and 12.7 to 19% for suicide ideation and suicide attempts respectively amongst school-aged adolescents between 14 and 18 years in the United States [11,12,13].

The data on suicide rates and patterns in Nigeria is limited, but the available studies have provided some insight into the issue. One study conducted on coroners' reports from 1957 to 1960 in the Western Region of Nigeria concluded that the suicide rate in Nigeria was low [14]. Another study conducted in hospitals between 1978 and 1981 found that the incidence of suicide attempts had not increased during that period, averaging a crude suicide attempt rate of 7 per 100,000. The most affected age group was teenagers, with less than ten suicide deaths per 100,000 [10]. A prospective study of self-harm cases reported in three major hospitals in Ibadan, Nigeria, found that 76.9% of the cases were under the age of 30, and 51.3% were students [15]. A study conducted by Omigbodun et al. aimed to determine the prevalence and associated psychosocial factors of suicidal ideation and attempts among young Nigerians. The study assessed 1,429 youths and found that over 20% of them reported having suicidal thoughts, while approximately 12% reported having attempted suicide in the preceding year, also of note is that more females reported suicide attempts [16]. The study highlighted that the rates of both suicidal ideation and attempts among young Nigerians were approaching the upper limit of acceptable rates. A study conducted in Lagos, a metropolitan city in Nigeria, found that 6.1% of individuals reported suicidal ideation, 4.4% reported suicidal planning, and 2.8% reported suicidal attempts within the past month [17]. According to Oladeji et al. [18], the 12-month prevalence of suicide ideation among adolescents was between 6.1–22.9% and 3–12.5% for attempts.

However, research on suicidal behaviour has received more attention in high-income countries, with few reports available in low- and middle-income countries, particularly in sub-Saharan Africa [19].

Unfortunately, data is not consistently available worldwide, especially in sub-Saharan Africa, where attempted suicide is a punishable offence [20, 21]. Documentation about suicide is sparse in Nigeria [20], Ghana [21], and Tanzania [5]. The stigma around suicidal behaviour makes estimating the prevalence of suicidal behaviour underestimated in these countries where suicide is illegal [22]. Additionally, religious, cultural, and social factors often influence the diagnosis, presentation, and reporting of suicides [23]. Hospitals may refrain from registering deaths by suicide, and death records are not stored, with no system in place to document the causes of death. As most data in developing countries is from hospital reports or police records, hospital-based surveillance is not suited to inform the evidence. Additionally, stigma and fear of harassment prevent relatives from disclosing the real cause of death [24].

There are several theories about suicide. The Interpersonal Theory of Suicide, introduced by Joiner in 2005, suggests that certain cognitive-affective states can develop due to stress from disrupted relationships, diminished social support, and overt rejection or exclusion leading to humiliation [25]. These states include perceived burdensomeness, which is the belief that one is a liability to others, and thwarted belongingness, which is loneliness and alienation. These states can lead to suicidal desires and attempts [26]. The Integrated Motivational Volitional Theory of Suicide (IMV) proposes that suicidal desires and attempts can be caused by stressful life events [27]. The Three-Step Theory of Suicide (3ST), introduced by O'Connor et al. in 2016, suggests that general life stress can contribute to suicidal ideation [28]. All theories indicate that having the capability to enact lethal self-harm is necessary for an attempt, which can be acquired through life events that cause pain, danger, or fear. These theories make it essential to continue to explore the coefficients of suicide [26].

Suicide is a hidden epidemic since the problems leading to it are often ignored or unnoticed [29]. The impact of suicide goes beyond just numbers and statistics. The guilt, grief, and pain experienced by friends and families are significant. There is also an alarming level of public ignorance about mental health services and suicide prevention measures [9]. The increasing suicidal tendencies among adolescents can be attributed to several factors, including increasing age [30, 31], poverty and limited opportunities [32, 33]. A person's socioeconomic status impacts their mental health [33, 34]. Risk factors for suicide attempts include depression [12, 31, 35], suicide ideations and plans [5], loneliness [12, 30], anxiety [31, 35, 36], unemployment, poverty [12, 37], bullying and other forms of violence [36, 38, 39], poor relationship quality with parents [40], low social support [41,42,43], and challenging living conditions [29, 33].

Suicide is a serious issue that affects not only individuals but also their loved ones. According to the WHO, there are over 20 suicide attempts for each suicide, which highlights the severity of this issue and its impact on young people and their families [3]. Despite being a significant public health problem, suicide prevention has not received adequate global attention.

There is insufficient research examining the prevalence of Suicidal behaviour comprising suicide ideation, plan, attempt, and actual suicide [6] their relationship with bully victimization, loneliness, and social support among Nigerian adolescents. A thorough understanding of the predictors of suicide ideation and attempts in the Nigerian adolescent population is critical for early suicide prevention. This study aimed to investigate the prevalence of suicidal thoughts, plans, and attempts among in-school adolescents in Benin City, Edo State, Nigeria. Additionally, it examines the relationships between suicide attempts, bullying victimization, social support, loneliness, sadness, and suicidal plans.

2 Materials and methods

The sample size for this study was calculated using a priori power analysis with G*Power software version 3.1.9.4 [44]. The study involved two groups and five response variables, with alpha set at 0.05, power (1–β) set at 0.80, and a small effect size f2 of 0.0225, giving a minimum sample size of 732. Seven hundred and fifty-six students were chosen from selected schools to ensure equal representation.

Data was collected using a self-administered questionnaire, which consisted of three sections. The first section was a semi-structured socio-demographic questionnaire that included questions on age, gender, family setting, and socioeconomic status. The respondents' socioeconomic status was determined using the Oyedeji classification of social class [45], which considers the parents' occupation and education level. The grading of social status ranged from 1 to 5, with one being the highest and five being the lowest. Several students reported being in the third socioeconomic class. The second section of the study used the Peer Relationship Questionnaire (PRQ) developed by Slee and Rigby [46] to determine victimization. The PRQ is a 20-item self-report questionnaire that measures bullying, victimization, and pro-social behaviour in children aged 12–18 years. It consists of three subscales and is scored on a 4-point Likert scale. The PRQ has a Cronbach α 0.70 [46]. For this study, a shorter version of the same questionnaire was used, consisting of 12 items and three subscales with four questions each. A score of five or below indicates non-victim [47]. The reliability coefficient for the present study was 0.72 for the victimization scale. The final section of the study included questions on anxiety, depressive symptoms, and a suicidality scale from Nigeria's 2004 Global School-based Student Health Survey (GSHS) Questionnaire [48]. The GSHS was developed by the World Health Organization (WHO) in collaboration with UNICEF, UNESCO, and UNAIDS, with technical assistance from CDC 2003 [48]. The survey contained nine questions about the mental health of respondents in the past 12 months. The dependent variable 'suicide attempts' was derived from one question in the GSHS: “In the past 12 months, how many times did you attempt suicide?” The answers range from “0” to “6 or more times”.

The data were dichotomized into reports of zero attempts (coded as 0) and one or more attempts (coded as 1). The study examined various factors such as demographics, behaviour, social environment, mental health, and family background. Independent variables were chosen based on previous research that linked suicide attempts to certain personal behaviours, demographic and family characteristics, and poor mental health. The survey included questions related to anxiety, loneliness, depressive symptoms, social support, and experiences of victimization to determine their associations. To assess anxiety, respondents were asked, "During the past 12 months, how often have you been so worried about something that you could not sleep at night?" with response options of "never," "rarely," "sometimes," "most of the time," or "always." We dichotomized the responses into "yes" (for "most of the time" or "always") and "no" (for "never", "rarely", or "sometimes"). To assess loneliness, respondents were asked, "During the past 12 months, how often have you felt lonely?" with the same response options. Again, we dichotomized the responses into "yes" and "no." To assess depressive symptoms, respondents were asked, "During the past 12 months, did you ever feel so sad or hopeless almost every day for 2 weeks or more in a row that you stopped doing your usual activities?" with response options of "yes" or "no." To assess social support, respondents were asked, "How many close friends do you have?" with response options of "0", "1", "2", and "3 or more friends." We interpreted "0" as indicating "no" support, while "1" and above were considered "yes" support. Nigerian authors have utilized different segments of this instrument [49, 50]. More details on creating these variables can be found in Table 1. Finally, the Peer Relationship Questionnaire was used to determine victimization experiences. The reliability coefficient for the victimization scale was 0.72.

Table 1 Independent variable derivation
Full size table

This study was a school-based cross-sectional descriptive survey. This study occurred in twelve secondary schools within a randomly selected local government area (LGA) of Benin City, Edo State. The participants were determined using a multistage random sampling technique. The names of the five local government areas that make up Benin City were written on separate envelopes and placed in a basket. An envelope randomly picked from the five in the basket indicated Egor LGA. We grouped the schools in the LGA into two based on ownership: public and private secondary schools. There were 24 senior secondary schools in the Egor local government area, 13 of which were public and 11 were private [51]. All public and private schools in Egor LGA was assigned numbers. The numbers were put in corresponding boxes for private and public schools, six schools from each group were selected using the simple ballot system. The number of students in each school was similar based on the calculated sample size. We informed all the students in senior classes about the study and gave a letter containing information about the study and a consent form. The students took home the documents to obtain consent from their parents or guardians and returned the completed forms. Those who met the inclusion criteria and returned the signed consent form participated in further screening for the study. Students picked a piece of paper with either a "Yes" or "No" option written on it from a box. The number of "No" options in the box varied depending on the total number of students who met the inclusion criteria in the class. We repeated the process for all the classes to have a total of 21 students per class and a total of 63 students per school. We replicated a similar process in the twelve randomly selected schools with 756 respondents. For each class, the student who picked the "Yes" option was put in a classroom within the school premises, free of interruptions/distractions from the teachers and students. We distributed questionnaires to the students. Since only 21 students filled out the questionnaire at once in a classroom, there was an opportunity for proper monitoring. 725 out of 756 respondents completed the questionnaire, which translates to a completion rate of 96%. The completed questionnaires were then analyzed. Two resident doctors from the community psychiatry unit were trained to work as research assistants.

The Ethics and Research Committee of the Federal Neuropsychiatric Hospital in Benin City approved the study methodology to ensure that it does not contravene international guidelines for research involving human subjects. We obtained informed consent from parents and caregivers, and the students gave their approval after explaining the aim and objectives of the study to them. Ethical issues, such as non-disclosure to others, the opportunity to decline an interview at any stage, and non-exposure to risk, were discussed with each respondent. The participants bore no financial burden for the study.

Our data was analyzed using IBM SPSS Statistics for Windows (version 25) predictive analysis software. The respondents' socio-demographic information was presented as descriptive statistics such as frequency, means, and standard deviation (SD). For categorical variables, we used chi-square (χ2) test for bivariate analyses. The results of the bivariate analysis were reported as frequencies,  proportions, chi square values with their statistical significance. We used logistic regression as a multivariate statistical technique, with suicidal attempts as the dependent variable and being a victim, depressive symptoms, anxiety, etc. as independent variables. The confidence interval was 95%, and all tests were two-tailed with statistical significance at a P-value of less than 0.05. The results of the regression analyses are documented using the Adjusted Odds Ratio (AOR).

3 Results

The study had a total of 725 participants, out of which 410 (56.6%) were male and 43.4% were female. The mean age of the participants was 15.23 years, with a standard deviation of 1.57. The study found that, during the recall period of 12 months, 13.5% of school-attending adolescents had suicidal thoughts, 11.4% had plans to commit suicide, and 10.5% had attempted suicide.

Table 2 shows the basic characteristics of students with suicidality (suicidal ideation, plans, and attempts), more males reported suicidality. These differences were not statistically significant (p-value: 0.33). The study found that many of the respondents who attempted suicide were in the middle adolescent age group, and this was statistically significant (p-value: 0.008). Furthermore, most of the parents of these adolescents (82.9%) belonged to socioeconomic class 3, and (72.4%) of individuals who attempted suicide also belonged to socioeconomic class 3 (p-value: 0.030). About a quarter of the respondents (24.6%) had witnessed an altercation between parents, but this was not statistically significant for suicidal attempts (p-value: 0.221). Also of note is that about half of those who reported suicidal behaviour had been victims of bullying behaviour and low social support.

Table 2 Basic characteristics of students with suicidality (numbers (n) and percentages (%))
Full size table

Table 3 describes the comparison of factors by suicide attempt among 725 school-attending adolescents in Benin City. As for birth order, 72.4% of the respondents were first children in their families, and although 77.6% of these individuals had made suicide attempts, it was not statistically significant (p-value: 0.305). The study also found that 48.7% of the respondents were victims of bullying in the past 30 days, and 32.8% of those who were victims of bullying had attempted suicide at least once (p-value: 0.050). In this survey, 17.1% of the respondents reported feeling lonely. Out of those who reported loneliness, 10.1% had attempted suicide. Still, this finding was not statistically significant (p-value: 1.000). An overwhelming majority of the respondents (88.28%) reported being anxious. Those who had attempted suicide were more likely to report anxiety (80.23%) (p-value: 0.022). In the past 12 months, 32.8% of the respondents had reported feeling persistently sad or hopeless for 2 weeks or more. Of those who had attempted suicide, 43.3% reported feeling unduly sad (p-value: 0.038). The percentage of respondents who had attempted suicide and reported having no social support was 30.67%, but this finding was not statistically significant (p-value: 0.46).

Table 3 Comparison of factors by suicide attempt among 725 school-attending adolescents in Benin City
Full size table

Table 4 presents the adjusted odds ratios (AOR) of covariates associated with suicide attempts, along with 95% confidence intervals (CI). The covariates are loneliness, gender, anxiety, undue sadness, suicidal thoughts and plans, social support and being a victim of bullying. The study revealed that those who had attempted suicide were more likely to report suicidal thoughts and planning (AOR = 11.32; 95% CI 6.70–19.14; AOR = 11.53; 95% CI 6.72–19.77) compared to those who had not attempted suicide. The respondents who had attempted suicide were also more likely to have been victims of bullying behaviour (AOR 0.62; 95% CI 0.380–1.008). They reported having symptoms of anxiety (AOR = 0.49; 95% CI 0.2653– 0.911) and feeling unduly sad (AOR = 1.66; 95% CI 1.0256–2.6938). However, in this study, gender (male) and lack of social support were not statistically significant at (AOR = 1.2769; 95% CI 0.783–2.080) and (AOR = 0.8234; 95% CI 0.4917–1.3786) with suicide attempts respectively.

Table 4 Outcomes of multivariate analysis of variables associated with suicide attempts among school-attending adolescents in Benin City, Nigeria
Full size table

4 Discussion

The study found that the rate of suicide attempts among in-school adolescents in Benin City, Nigeria, was 10.5%. This figure is lower than the rates documented for adolescents of similar age in other sub-Saharan African countries, which varied between 17.5% and 31.1% [12, 16, 30, 35]. The rate is also lower than that found among adolescents in high-income countries, which was 15% and 12.7% to 19.0% in different studies [13, 22]. The relative social and political stability of Benin City compared to other regions in Nigeria may partly explain the lower rate of suicide attempts. Other studies have suggested that conflict, volatility, and uncertainty negatively affect psychological well-being [41]. However, it is worth noting that the study only sampled adolescents within a local government area in Benin City, which means the findings may not be representative of the entire adolescent population in Edo State and Nigeria. The difference in residences between urban and rural environments may also account for variations in the rates of reported suicidality [33]. Additionally, socioeconomic, educational, and social-cultural factors may be responsible for the differences between rural and urban rates of suicide attempts [33]. It is also possible that the lower rates of suicide attempts found in this study reflect cultural myths and taboos against suicide in Benin City [24]. The study found a higher number of males involved in suicidality when compared to females. This contradicts the reports by authors in the region [12, 30] and globally [52], who found that females were more involved in suicidality. However, there was no significant association between gender and suicide attempts [30, 53]. This study suggests that this finding may reflect a skewed underreporting of suicidality, which may not allow for gender differences in suicide attempts reported. This is like a study in Nigeria, which utilized the Diagnostic Interview Schedule for Children and reported no gender-specific interaction with suicidal ideations or attempts [16]. This may be due to the multifaceted interactions among social factors in the country, which are sufficiently raised to confound gender differences. It is important to note that more male students were in the study population than female students. This may be because some parents believe that girls belong in the kitchen and that formal education is not for them [54].

The study also found that loneliness, anxiety, depressive symptoms, suicidal thoughts and suicidal planning remained significantly associated with attempted suicide in the multivariate analysis. These findings are consistent with studies of suicidal behaviours in high and low-income countries that have shown loneliness [12, 30], depression [12, 35], anxiety [35, 36], suicidal thoughts [5] and plans [5] have associations with increased suicide attempts.

According to our study, 48.69% of participants reported experiencing violence, which is lower than the rates reported in other regions of the country [38]. This could be because Benin City is more urban than the areas examined in previous reports. The victims of violence in this study had a higher tendency to attempt suicide. This is consistent with another study that found no specific role in bullying to be the most significant predictor of suicidal thoughts and actions. All bullying roles: victims, bullies, and those who fall into both categories, have all been associated with an increased risk of contemplating or attempting suicide [39]. The findings of the study reported no significant relationship between social support and suicide attempts. Contrary to the reports in the literature, adolescents who reported having no social support (no friends) were more likely to have attempted suicide [41, 42]. The variance may be due to misunderstanding the word "close." The respondents possibly interpreted the word "close" to mean bosom or a confidant rather than friends.

It has been found through multivariate analyses that middle-aged adolescents have a higher risk of attempting suicide. Despite studies conducted locally [30] and globally [31] that link increasing age with a greater propensity towards suicide attempts, the findings here may be influenced by an underreporting bias among older adolescents, who are more likely to be swayed by cultural taboos and myths against suicide. While late adolescents should report higher suicide attempts, more adolescents were found to be in the middle adolescent age group compared to the late teenage age group.

5 Limitation

This study provides valuable insights into adolescent suicidal behaviour in this area. However, there are some limitations to the study that must be considered:

  1. 1.

    The cross-sectional nature of the study makes it difficult to determine causal relationships between suicide attempts and associated factors.

  2. 2.

    The study sample is representative only of school-age adolescents attending school in one local government area in Benin City. Future research should include a more representative sample of out-of-school adolescents to validate the findings and explore other factors associated with suicidality among all adolescents.

  3. 3.

    The 12-month recall period in the survey questions may lead to recall bias.

  4. 4.

    The GSHS is a survey instrument designed for use across cultures, which may lead to some ambiguities in the interpretation of questions by respondents depending on the cultural presentation of ailments like depression, anxiety, and loneliness. The development of standardized and culturally relevant scales in our culture should be encouraged.

  5. 5.

    Underreporting bias cannot be excluded because of taboos and myths surrounding suicide.

6 Recommendation

Our research unequivocally demonstrates that mental health issues such as depression, anxiety, and loneliness are significant contributing factors to suicide attempts. We recommend that suicide screening should be encouraged in all clinics. Age-appropriate mental health promotion programs for children and adolescents should be of utmost priority in sub-Saharan Africa. Decriminalizing suicide is paramount to encourage individuals to seek help without fear of being arrested and create a safer environment. Legislation must inform prevention and intervention strategies to reduce childhood suicide rates. Finally, future research must immediately prioritize identifying variables that moderate suicide behaviours in Nigeria.