Mandatory Junior School Certificate Exams and Young Teenage Suicides in Bangladesh: A Response to Arafat (2020)

We recently published the short article ‘Young teenage suicides in Bangladesh – are mandatory Junior School Certificate exams to blame?’ (Mamun and Griffiths 2020a). In response to this, Arafat (2020) made three critical claims. In our view, each of these claims is unfounded and/or simply incorrect. Arafat’s first claim: “Bangladesh has no national suicide database...Therefore, there is no data to support the statement that after starting the mandatory junior school certificate exam, the suicide rate has been increased” (p.1). Our response: We are only too aware that Bangladesh has no official suicide surveillance or database system and have highlighted this very fact in a number of our previous articles (e.g., Mamun and Griffiths 2020a, b, c; Mamun et al. 2020a). We never once claimed on our article that the Bangladeshi suicide rate had increased as a consequence of the introduction of the junior school certificate (JSC) exam. Arafat has highlighted something that was never there to start with. We simply reported that there had been (at least) 39 teenage suicides reported in Bangladesh media stories during a 3-year period (2017–2019). We also speculated that the number of suicides among this group may be higher than other educational cohorts and provided the example of a Bangladeshi medical sciences student cohort. More specifically, we stated:

We simply noted that using the same methodology, there were 13 cases of suicide among a 6year cohort of Bangladeshi medical students in a 23-month period compared with 39 cases of suicide among a 1-year cohort of Bangladeshi school teenagers in a 36-month period. We simply compared the incidence of teenage suicide with the only previous comparable study of student suicides existing in the country (i.e., medical sciences-related student suicides). The incidence of suicide among Bangladeshi school students is clearly much higher than among Bangladeshi medical students. We were very careful in the language we used and deliberately used words such as "may" and "suggests" rather than anything that was definitive.
Arafat's second claim (part 1): "Repeated studies have been failed to find out anyone to one relationship [sicwe assume Arafat means 'any one-to-one relationship'] with any risk factors and suicide" (p.1).
Our response: In the response of the second issue, we do not disagree that there are multiple risk factors for suicide. However, risk factors for suicide are not the same as reasons for suicide. In the press reports we collated, the main reasons attributed to the majority of the 39 suicides were related to JSC-academic distress (i.e., failing the JSC and/or not getting highest JSC grade). This appears to have led to mental suffering (e.g., depression, anxiety) and poor mental health is one of the key risk factors for suicide (Ang and Huan 2006;Arafat and Mamun 2019;Jayanthi et al. 2015;Mamun et al. 2020a, b).
Arafat's second claim (part 2): "Suicide is the output of complex interactions of multiple factors mentioned in multiple ways such as the interaction of proximal and distal factors, interaction nature, and nurture" (p.1).
Our response: While this statement is broadly true, there is growing worldwide evidence of 'impulsive' suicides with or without facilitating multiple risk factors (Auerbach et al. 2017;Cha et al. 2018). Impulsive suicides are not unusual in Bangladesh and Arafat has reported this in his own studies. For instance, in one study, 81% suicidal ideators reported their ideation to be impulsive (Arafat et al. 2018a), and in another study, 63% of individuals attempting suicide reported having no prior suicidal thoughts and 58% of attempted suicides were reported as being impulsive (Roy et al. 2018). Impulsive suicides have also been reported in Bangladesh and India (Arafat and Hossain 2018). We were therefore surprised that we were criticized given Arafat himself has reported in a number of his papers that suicide can be an impulsive behavior.
Arafat's third claim (part 1): "…the authors used the data from the contents of newspapers which cannot be such strict [sic] to conclude anything…Previous studies assessing the quality of media reports of Bangladesh found that they are poor when compared with the World Health Organization reporting guidelines" (p.1).
Our response: We were frankly astonished to read this claim given that Arafat has published at least seven studies using the exact same methodology we used (i.e., Arafat et al. 2018b;Arafat and Hossain 2018;Arafat et al. 2018bArafat et al. , 2020aArafat and Mamun 2019;Shah et al. 2017). A lot of Arafat's empirical data on Bangladeshi suicides utilizes media reports. The reason we collated media reports is probably the same reason that it is Arafat's preferred method (i.e., there is no national suicide database in Bangladesh). Published papers using data from media reports are often used in countries where there is no current and/or active national suicide database available such as Bangladesh, India, and Pakistan (Armstrong et al. 2019;Bhuiyan et al. 2020;Dsouza et al. 2020;Mamun et al. 2020a, b;Mamun and Griffiths 2020c, d; Ullah 2020) (for a systematic review on global media reporting suicides, see Niederkrotenthaler et al. 2020). Arafat says that the quality of media reports in Bangladesh is poor based on the WHO reporting guidelines. The WHO criteria provide 12 "dos and don'ts" when it comes to reporting on suicide in the media. Journalists are advised: "(i) do provide accurate information about where to seek help, (ii) do educate the public about the facts of suicide and suicide prevention, without spreading myths, (iii), do report stories of how to cope with life stressors or suicidal thoughts, and how to get help, (iv) do apply particular caution when reporting celebrity suicides, (v) do apply caution when interviewing bereaved family or friends, (vi) do recognize that media professionals themselves may be affected by stories about suicide, and (vii) don't place stories about suicide prominently and don't unduly repeat such stories, (viii) don't use language which sensationalizes or normalizes suicide, or presents it as a constructive solution to problems, (ix) don't explicitly describe the method used, (x) don't provide details about the site/location, (xi) don't use sensational headlines, (xii) don't use photographs, video footage or social media links" (WHO 2017; p.viii).
Given that the WHO reporting guidelines have so many criteria to adhere to, almost every media report in (and outside of) Bangladesh would be rated as "poor." Media reports typically report the method of suicide, include a photo of the individual who committed suicide, have the word 'suicide' in the title, and typically provide less information about suicide support services. The basic information we extracted from the media reports were generally aspects that have little to do with the WHO criteria (i.e., city, gender, age, and reason for committing suicide). Arafat claims we cannot "conclude anything" based on such media reports but using the basic facts reported, we can and we did.
Arafat's third claim (part 2): "They [press reports] regularly mentioned the details of victims, methods of suicides, life events, and monocausal explanations…Authors [sic] also tried to explain the similarity between teenage suicides and medical students' suicide which could be flawed based on this study. These are two situations, two life stages, two academic stages" (p.1).
Our response: The similarity was in the methods used to collect the data (a method which Arafat himself commonly uses) and the fact they were both student populations. The fact that they are two different cohorts at different stages in their academic career is arguably a positive (not a negative) because many of the medical student deaths were also due to academic distress and failure. The notion that our comparison is somehow "flawed" is arguably nonsense given we simply made comparisons with the only other published data and they showed some similarities. Most media reports provide monocausal explanations of suicide cases and most of the reasons tend to be proximal. As noted above, the reasons given for suicide are not the same thing as risk factors of suicide and Arafat clearly conflates (or confuses) the two. There is no way of identifying the actual suicide risk factors among suicide victims because they are deceased (Shain 2007). Therefore, coroners often utilize a psychological autopsy (Isometsä 2001) in which other individuals (e.g., family members, friends, relatives, police or health care professionals) are interviewed. Journalists often use similar methods to identify the reasons underlying the suicides that are reported.
Arafat alleges that our study was "flawed" and of "poor scientific quality" (p.1) mainly based on the methodology we used. Given that Arafat and his colleagues have published many studies utilizing exactly the same method we used, readers could perhaps conclude (using his own argument) that his own body of work in this area is similarly flawed and of poor quality.
The final (and most important) point to make is that the title of our study deliberately ended with a question mark. Our study raised a question and we simply noted that in the media reports of the majority of the teenage suicides, the reason for the suicide highlighted was JSCrelated. Arafat said our claim was "hypothetical" (which is defined as "imagined or suggested but not necessarily real or true" [Cambridge Dictionary 2020]). The information in the media reports that we extracted were neither imagined or suggested. They were reported and we faithfully reproduced what we found. We are only too aware of the limitations of media reports (as is Arafat given it is his preferred method of data collection) but that does not mean we cannot ask the question.

Compliance with Ethical Standards
Conflict of Interest The authors declare that they have no conflict of interest.

Informed Consent Not applicable
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.