Comment on ‘Posttraumatic stress disorder and depression after the 2018 Strasbourg Christmas Market terrorist attack: a comparison of exposed and non-exposed police personnel’: moving towards shared methodology in terrorism-research

ABSTRACT

In the recent article 'Post-traumatic stress disorder and depression following the 2018 Strasbourg Christmas Market terrorist attack: a comparison of exposed and non-exposed police personnel,' important insights are provided about the association between terrorism exposure and Post-Traumatic Stress Disorder (PTSD). However, in our opinion, there are several methodological issues that limit the results of this study. In this letter, we discuss the problematic use of partial PTSD, the adjustment of the PCL-5 to refer only to a specific terrorist attack, and the significance of criterion A.

En el reciente artículo ‘Trastorno de Estrés Postraumático y Depresión tras el Ataque Terrorista al Mercado Navideño de Estrasburgo de 2018: una comparación del personal policial expuesto y no expuesto’, se brindan importantes conocimientos sobre la asociación entre la exposición al terrorismo y el Trastorno de Estrés Postraumático (TEPT). Sin embargo, existen varias opciones metodológicas que, en nuestra opinión, limitan los resultados de este estudio. En esta carta, discutimos el uso problemático del TEPT parcial, el ajuste de la PCL-5 para referirse sólo a un ataque terrorista específico y discutimos la importancia del criterio A.

在最近的文章‘2018 年斯特拉斯堡圣诞市场恐怖袭击后的创伤后应激障碍和抑郁：暴露和未暴露警察人员的比较’中，提供了关于恐怖主义暴露与 PTSD 之间关系的重要见解。然而，我们认为有几种方法选择限制了本研究的结果。在这封信中，我们讨论了部分 PTSD 的使用存在问题，将 PCL-5 调整为仅指一次特定的恐怖袭击，并讨论了标准 A 的重要性。

KEYWORDS:

• Post-traumatic stress disorder
• terrorism
• criterion A

PALABRAS CLAVE:

• trastorno de estrés postraumático
• terrorismo
• criterio A

关键词:

• PTSD
• 恐怖主义
• 标准A

We read the study by Nourry et al. (2023) with great interest. However, we would like to address two methodological issues that are recurring challenges in the field of terrorism studies (Durodié & Wainwright, 2019).

1. The concept of partial PTSD

Firstly, we have some concerns regarding the concept of ‘partial’ PTSD. As the name suggests, partial PTSD refers to individuals who exhibit some symptoms of PTSD but do not meet all the criteria for a positive screening. More specifically, the authors define partial PTSD as: ‘Partial PTSD was considered when a participant met two or three of the diagnostic criteria B, C, D, or E’ (Nourry et al., 2023). These diagnostic criteria B to E are based on the DSM-5 definition of PTSD (American Psychiatric Association, 2013).

There are two issues with the use partial PTSD. First, many studies have used another definition of partial PTSD, namely one where partial PTSD must meet symptom criterion B and at least one other symptom criterion (e.g. Pietrzak et al., 2021; Schnell et al., 2020). Indeed, partial PTSD is not defined in the DSM (Fischer et al., 2023), meaning that there is no ‘correct’ way of approaching partial PTSD. However, it does hamper the comparability of the findings regarding partial PTSD.

Second, another issue in the current study is that partial PTSD is not used as its own diagnostic category, as partial PTSD and full PTSD are merged together into one group, called ‘PTSD’. So, by grouping both partial and full PTSD together, it defeats the purpose of examining risk factors for partial PTSD and for PTSD, as it is unclear whether or not the risk factors are associated more for those with partial PTSD, or with those with PTSD. For example, the authors state ‘The participants with partial or complete PTSD used mental health care services significantly more than those with no PTSD (33% and 6%)’ (Nourry et al., 2023), without distinguishing partial and full PTSD. Yet, it would have been interesting to know if those with partial PTSD seek out mental health care services more than those with PTSD, as some studies suggest that symptom severity after terrorist attacks is associated with unmet mental health aid needs (Ghuman et al., 2014).

2. The use of PCL-5 and criterion A

The second methodologic issue is related to criterion A, which might not be met for some participants. Nourry et al. (2023) state that: ‘All respondents were asked to refer to the Christmas Market terrorist attack as a potentially traumatic event’, thus implying an adjustment of the PCL-5 by solely focusing on the specific terrorist attack. This raises doubts about the validity of the scale, as respondents who may have had PTSD prior to the attacks are forced to attribute their symptoms exclusively to that specific event.

Many respondents in the study were not directly exposed to the event referenced by the scale. According to the study's inclusion criteria, all police personnel employed by the Strasbourg police departments in December 2018 were eligible for participation, including administrative, technical, and scientific staff. The authors even state, ‘We expected that at least half the police personnel in our sample would have been unexposed to the attack, either because they were not working on the day of the attack or because of their assignment’ (Nourry et al., 2023). This raises questions about whether criterion A of PTSD is fulfilled for all participants in the study.

Criterion A serves as the ‘gatekeeper’ for PTSD diagnosis, requiring exposure to ‘death, threatened death, actual or threatened serious injury, or actual or threatened sexual violence’ as outlined in the DSM-5 (American Psychiatric Association, 2013; Durodié & Wainwright, 2019). One way of meeting this criterion is through ‘indirect exposure to aversive details of the trauma, usually in the course of professional duties (e.g. first responders, medics)’ (American Psychiatric Association, 2013). Indeed, even with indirect exposure, the psychological impact of the aftermath of a terrorist attack can be severe (Skryabina et al., 2021). However, an issue in the current study is that the ‘non-exposed’ group is defined only in the sense that they are not part of the ‘exposed’ group. Yet, considering that this group constitutes 44.6% of the entire sample, the lack of information on what ‘non-exposure’ is, makes it difficult to assess how they might have completed the PCL-5. Especially if we consider that most of these non-exposed were technical and administrative staff of whom it is already somewhat unclear if they could be assumed to be in a situation to meet criterion A.

We do not dispute that respondents who screened positive for PTSD or partial PTSD may indeed have mental health issues. It can be argued that allowing the respondents to choose the index-event is better, as criterion A would be clearly met. After all, considering the high number of traumatic events in the sample, we think it would probably not have been difficult for respondents to fit criterion A with their own index-event. Furthermore, some of these respondents might already have had partial PTSD or PTSD before the event, considering their trauma-exposure.

3. Conclusion

Both combining partial PTSD and PTSD, and not having certainty of the criterion A-event are improper use of the DSM-5 PTSD criteria. Utilizing such a loose interpretation of these criteria will inflate PTSD rates, as has been observed in numerous previous studies (Durodié & Wainwright, 2019). Also, the comparability between studies is compromised and prevalence rates become less meaningful when people with different severities and diagnoses are grouped together. For instance, those who were directly exposed and feared for their life is categorized alongside those who were not even present during the event. This is not to say the mental health issues of people with partial PTSD are to be ignored. However, terrorist victims with mental health issues have indicated that the overutilization of mental health diagnoses deterred them from seeking help (Van Overmeire et al., 2021).

Thus, we strongly advocate to restrict DSM screening criteria for PTSD to their intended use for the sake of harmonization across studies and avoiding overestimation of prevalence rates.

Disclosure statement

No potential conflict of interest was reported by the author(s).