Prevalence of depression, anxiety and associated factors among school going adolescents in Bangladesh: Findings from a cross-sectional study

A cross-sectional and self-reported survey was conducted among adolescent students aged 13–18 years at selected schools (secondary and higher secondary) in Dhaka City (in Bangladesh) from November 2019 to February 2020. A multi-stage sampling technique was employed to recruit subjects for this study. In the first stage, a number of schools (primary sampling unit) were selected, conveniently located in the study area. In the selected schools, each grade is divided into several sections. Then, sections were randomly selected and students in the selected sections were then eligible to participate in the study. Overall, six schools including four secondary (grades 8–10) and two higher secondary (grades 11–12) schools, (24 classes) were included in the study. Inclusion criteria for participation included adolescent age range and willingness to enroll in the survey. Exclusion criteria were not completing the full survey.

The sample size was calculated using the following equation: n=Z2pqe2×de;n=1.962×0.366×(1−0.366)0.052×1.5=534.9≈535 Here, n = number of samples; z = 1.96 (95% confidence level); p = prevalence estimate (.366); q = (1-p); e = margin of error; de = design effect for multistage sampling.

In the most recent study, Anjum et al. (2019) reported a prevalence rate of depression of 36.6% focusing on urban and semi-urban adolescents in Bangladesh [11]. For multi-stage sampling, the margin of error and the design effect are usually considered to be 5% and 1.5, respectively. In the present study, the calculated sample was to be 535. However, 563 participants were recruited to ensure adequate power for the study.

Data were analyzed using Microsoft Excel 2010 and SPSS Statistics version 25. Microsoft Excel was used for data entry, editing, and sorting. Finally, an excel file including all variables was imported into SPSS software. Descriptive statistics (e.g., frequencies, percentages, means, standard deviation, etc.) and first-order analysis (i.e., chi-square tests) were calculated with SPSS software. Logistic regression (both unadjusted and adjusted models) was performed, yielding odds ratios and their 95% confidence intervals, to examine associations between categorical dependent and independent variables. Analyses were univariable, yielding crude odds ratios (COR), followed by multivariable analyses with only significant factors (from univariable analyses) entered together, with the exception of anxiety and depression in the models of each other, yielding adjusted odds ratios (AOR). The association of variables was considered statistically significant if the two-sided p-value was less than .05.

All procedures of the present study were performed in accordance with the Intuitional Research Ethics board. The study was approved by the Biosafety, Biosecurity, and Ethical Clearance Committee, the ethical review board of the Faculty of Biological Sciences, Jahangirnagar University, Savar, Dhaka-1342, Bangladesh. Formal permission was taken from the respective school/college authorities. Class teachers read the questionnaire and approved it to collect data prior survey. The written consent was obtained from each participant prior to data collection. The informed consent documented clearly explaining the aims and procedures of the study, risks, and benefits associated with participation, their right for voluntary participation and their right to withdraw from the study. Moreover, parents/guardians were also informed about the aim of the study through an ascent form. Participants’ information like name, mobile number, and address was not recorded, and coding was used in the questionnaires. Anonymity and confidentiality were ensured for all participants.

In the present study, the findings indicated that nearly one-third of respondents (26.5%) experienced moderate to severe depression. Compared to previous studies in Bangladesh among adolescents, the prevalence of depression in the present study was higher (26.5% vs. 25%) than a previous study conducted in 2013 among 898 adolescents aged 13–16 years in Dhaka City [28]. The prevalence of depression in the present study was lower (26.5% vs. 36.6%) than a study conducted in 2018 among 311 adolescents aged 13–17 years in Dhaka (urban and semi-urban areas) [11] and also lower (26.5% vs. 49%) compared to a previous study conducted in 2012 among 165 adolescents aged 15–19 years selected from two urban schools in Bangladesh [29]. In the present study, one-fifth of respondents (18.1%) experienced moderate to severe anxiety. Direct comparison of this finding is difficult due to a lack of similar Bangladeshi studies.

Table 3 demonstrates the prevalence rates of depression and anxiety among adolescents in Asian jurisdictions. The rate of depression and anxiety across all these studies likely varies in relation to a number of factors. First, given that depression and anxiety represent a spectrum of severity, differences in cut-offs occasioned by different instruments likely will yield differences in prevalence estimates. Second, given strong contextual effects in common mental disorders, rates will vary also across differences in socio-demographic settings, age groups, and age-period-cohort strata.

Significant associated factors of depression were having unsatisfactory sleep, being a cigarette smoker, and having anxiety (see Table 2). The findings revealed that participants with unsatisfactory sleep were associated with depression which was similar with previous reports in adolescent populations in Bangladesh [11] and China [31]. It is also consistent with a large-scale study in Norway [32]. This finding also supports a recent case-control study conducted with adolescents [33]. The present report also revealed that adolescents who were currently smoking had greater odds of depression compared to those who did not smoke. This finding is in line with previous reports [34–35] and also consistent with a randomized controlled trial [36]. However, the present study found no statistically significant associations between depression among adolescent students and father’s education and occupation, mother’s education and occupation, living with parents, family size, and school grade.

The present study’s findings also demonstrated that significant factors of anxiety were being older age (15–16 years), not having a good perceived relationship with friends, and having depression. The present report reflected that older adolescents were more prone to being anxious which is in line with a presenting study conducted in Jordan [2]. The present findings revealed that not having a good perceived relationship with friends had greater odds of anxiety compared to those who had a perceived good relationship with friends. This is concurrent with a prior report which found that the perceived care by friends mediated the effect of making friends on social anxiety [37]. The same study concluded that caring relationships with friends and parents each play a role in protecting early adolescents against increasing levels of social anxiety over time. Another report also found that anxiety was associated with self-reported impaired friendship quality [38].

The present study revealed a strong association between anxiety and depression. The assessment of anxiety in the context of depression, and vice versa, is important, given frequent comorbidity with depression and the resulting diagnostic and clinical complications [39]. Anxiety and depression are closely related nevertheless may represent separate domains [40, 41], particularly in adolescents [42]. As expected, anxiety was highly comorbid with depression, such that the anxiety comorbidity rate was 76.5% for depression and the depression comorbidity rate was 41.1% for anxiety. Anxiety comorbidity in depression, and vice versa was also previously found in Bangladesh among different populations [43–45]. This is relevant, as comorbidity negatively affects severity and prognosis [46–48] and early anxiety may increase the probability of occurring depressive disorder [49]. Therefore, from a public health perspective, it is important to focus on both anxiety and depression in combination.

A notable difference was the association between depression and male sex obtained in the univariable analysis. While unusual, such differences have been reported previously in cross-sectional studies [12, 50], and should be interpreted as a function of the local mix of socioeconomic and cultural factors impacting differentially on depression in men and women [50]. Indeed, in the multivariable analysis, male sex was no longer associated with depression, suggesting the sex difference was reducible to other factors that may be specific to Bangladeshi culture in terms of differential impact on sex.

Other factors that were not examined but which may have contributed to the relatively high prevalence rates and sex difference include ethnicity, religion, parental education level, type of housing, alcohol intake, and mental health status of parents [2, 6, 7].

The present study should be interpreted in the light of several limitations. Firstly, the study used self-report methodology that may have prejudiced the outcomes through well-known factors such as memory recall dependence and social desirability. Secondly, the study was cross-sectional in nature which hampers causal attribution. Carrying out a longitudinal study would overcome this limitation in understanding potential causal relationships. Thirdly, the modestly sized sample is only representative of selected schools (using convenient sampling) of Dhaka City in Bangladesh and might not generalize to other Bangladeshi schools or schools in other countries. Therefore, studies employing larger samples from more representative school populations are required. Moreover, the present study used psychometric scales that were previously used/ validated among the general population in Bangladesh rather than adolescent samples. Likewise, the cut-offs of the screening tools were not well established for Bangladeshi adolescents against gold-standard assessments. The exclusion of adolescents who might have dropped out of school also limited the study findings. Finally, the factor ‘sleep’ was included as an associated factor but also may represent a symptom of anxiety and depression. Nevertheless, sleep was included separately because it is a transdiagnostic factor with major functional consequences but also can occur in isolation representing an important associated factor for a range of mental disorders. In other words, it can be both parts of the disorder and a symptom of the disorder.