Anxiety, Depression and Posttraumatic Stress Disorder after Terrorist Attacks: A General Review of the Literature

Terrorism, though not well-defined, is a violent act that has been shown to have longstanding effects on the mental health of those who witness it. The aim of this general literature review is to explore the effect that terrorism has on posttraumatic stress disorder (PTSD), major depressive disorder (MDD) and anxiety disorders, as well as the bio-psycho-social determinants that mediate its impact. This paper describes the prevalence, risk factors, protective factors, common presentations and interventions identified for PTSD, depression and anxiety disorders occurring following terrorist attacks. We conducted a literature search in MEDLINE using a number of keywords detailed below. After applying inclusion and exclusion criteria, we kept 80 articles, which we summarized in tabular form. A majority of articles found detailed the impact of terrorism on PTSD, and took place in a Western, mainly American setting. The main factors that impacted the presentation of mental illness include gender, ethnicity, social supports, socioeconomic status, level of preparedness, level of exposure, pre-existing trauma and mental illness, and subsequent life stressors. The main intervention detailed in this article as showing evidence post-terrorism is trauma-focused cognitive-behavioural therapy. This study highlights the importance of this topic, and in particular, its implications for public health policy and practice.


Introduction
There is no widely agreed upon definition of terrorism. In general terms, it is defined as an act of violence that is used to further a political goal by instilling fear into the public [1,2]. Though acts that could be defined as "terrorism" have occurred since the Roman Empire, the term was coined during the French Revolution [2]. Since the terrorist attack in New York City on 11 September 2001 (9/11), it is difficult to imagine a topic that has shaped global discourse as much as terrorism. In 2019, there were over 8300 terrorist attacks worldwide with about 25,000 fatalities from terrorism [2]. Over the course of the 21st century, fear of a terrorist attack occurring in proximity has grown, especially in the Western world. In 2018, over a third of individuals surveyed in countries such as Spain, Great Britain and Germany believed there would be an attack on home soil [3].
Academic interest in the association between terrorism and health, namely mental health, started in the late 90s. However, interest peaked following the 9/11 attacks. Most research centered around the newly christened diagnosis of posttraumatic stress disorder (PTSD) in the DSM-IV [4]. More than a decade later, research has also been done regarding the burden of other mental health conditions following a terrorist attack, such as major depressive disorder (MDD) and anxiety disorders; though to a lesser extent than PTSD [5]. Although there is an abundance of information and data on PTSD, anxiety disorders and MDD in other contexts, we are specifically curious regarding research in the context of terrorism.
This article presents a review of the recent literature on the topic. In this paper, we aim to describe the prevalence, risk factors and protective factors related to the development of 9.7% had current, 7.9% remitted, and 5.9% partial PTSD. Avoidance and hyperarousal symptoms were most common, and flashbacks least common. PTSD was associated with health and well-being, mainly dissatisfaction with life.
Chen, C. et al., 2020 [10] The Risk factor for PTS was both direct, and media-based (live television) exposure to the attacks 5-6 years later. Psychiatric disorders after terrorist bombings among rescue workers and bombing survivors in Nairobi and rescue workers in Oklahoma City Event, date & place: the 1998 U.S. Embassy bombing in Nairobi and the 1995 Oklahoma City bombing Type of study: comparative study N: 52 male rescue workers (Nairobi), and 105 directly exposed male civilian survivors (Nairobi) and 176 male rescue workers (Oklahoma) Methods: the Diagnostic and Statistical Manual of Mental Disorders-IV was used to assess pre-disaster and post-disaster psychiatric disorders, and variables related to demographics, exposure, disaster perception and coping was gathered.
22%: PTSD in Nairobi rescue workers and civilians 27%: MDD in Nairobi rescue workers and civilians The prevalence of PTSD and major depressive disorder after the Nairobi attack was 2-4 times higher than among rescue workers in Oklahoma City.  TV viewing and decreased perceived safety were related to increased PTSD and major depressive disorder. High perceived threat was associated with PTSD among those directly and indirectly exposed.  In individuals who were indirectly exposed, social network use and dysfunctional emotion regulation strategies were found to be risk factors for increased emotional distress. Type of study: retrospective study N: 17 highly exposed survivors Methods: the relationship between PTSD symptomatology, a common genetic polymorphism of the serotonin transporter and neural activity response was assessed using functional MRI while showing images associated with 9/11 and images not associated as a control.
Carriers of the short allele had higher levels of PTSD. Posterior cingulate cortices activity mediated the relationship between the genotype and PTSD.  Type of study: longitudinal study N: 531 mothers and fathers of exposed children Methods: assessed at 5 and 14-15 months following the Utoya shooting, their rates of PTSD, anxiety disorders and major depressive disorder were compared to age and gender adjusted expected scores of the same illness in the general population. Parents of exposed children had 3 times higher rates of major depressive disorder and anxiety disorders than the general population, and rates of PTSD were 5 times higher.  Optimism may help to neutralize the effects of high exposure on levels of symptoms of avoidance, numbing, and dysphoric arousal but not on the symptoms of intrusions and anxious arousal Employees who reported extra effort displayed increased risk for PTSD rates. Perceived reward for extra effort from a leader was associated with lower risk for PTSD. There was no statistical difference in the prevalence of mental illness between the two groups. Auditory reminders were most frequently encountered. most distressing and highly associated with meeting criteria for PTSD. 6.3%: PTSD Life stressors and premorbid mental illness leads to high symptoms presentation. Individuals in the high symptoms category most frequently screened positively for depression and functional impairment.  Longer duration of untreated illness (202 vs. 92 months) and higher symptom severity were seen in the terrorism group. Trauma-focused cognitive behavioral therapy is efficacious and useful in clinical practice for the treatment of PTSD in victims of terrorism. An intention to treat analysis was performed to assess its efficacy.
At post-treatment and at the 1, 3, 6 months, and 1 year follow-ups, large statistically and clinically significant decreases in PTSD, MDD and anxiety disorders were found using trauma-focused cognitive behavioral therapy. Being "very dissatisfied" with debriefing was associated with symptoms of avoidance and numbing, as well as higher rated of PTSD. Type of study: comparative study N: 55 directly exposed emergency responders (37 underwent debriefing, 18 did not) Methods: assessed outcomes between the two groups using the PHQ-9, WHOQOL-BREF, PCL and BSI.
Lower quality of life and increased major depressive disorder were found in the group who had undergone crisis intervention.

Prevalence and Risk Factors
Post-traumatic stress disorder is an outcome that is commonly explored throughout research on terrorism. In the general American and European population, 1-year prevalence is between 0.9-3.5% [84]. The articles featured in this review identify a higher prevalence of PTSD in individuals both directly and indirectly exposed to terrorist attacks. Given the increase in terrorism research that occurred post 9/11, the majority of the articles featured in this review assessed those populations. In direct survivors, the prevalence of PTSD has been found to be around 30% [74]. It reaches approximately 39% in the first 6 months, and slowly decreases to about 22% after 1 year [74]. The prevalence of PTSD without any comorbidities in survivors was found to be about 4.1%, even 14-15 years after the attack [54]. In traditional relief workers, such as first responders and rescue workers, prevalence for PTSD is lower in the first 3 years, and slowly climbs up to 10%, peaking approximately 5-6 years post-attack [10,73]. In non-traditional relief workers, such as volunteer workers, rates of PTSD are much higher, climbing to 21.9% [10]. The prevalence is about 23% in relatives or close friends of victims who were injured or killed in terrorist attacks [74].
Though there is not an abundance of articles examining the mental health effects of terrorism set in countries other than the U.S.A., some of these have been reviewed for this article. Certain communities showed higher proportions of PTSD, others lower. However, it can be challenging to compare the prevalence rates given that the level of exposure of the populations studied is heterogenous. In Nairobi, the prevalence of PTSD in survivors and rescue workers following the 1998 U.S. Embassy bombing was 22% which was found to be 2-4 times the rates following the Oklahoma City bombings [26]. In the 4-6 weeks following the Bardo museum attack in Tunis in 2015, one study found that 68.6% of museum works displayed posttraumatic stress symptoms [13]. Similarly, 5 months following the Qissa Khwani Bazaar bombing in Pakistan, 77% of direct survivors suffered moderate to severe PTSD [6]. Following the 2015 Ankara bombings in Turkey, one study found that PTSD prevalence in direct survivors was 24.7% [12]. In contrast, following the 2011 Oslo bombing, only 2% of trained professionals and 15% of unaffiliated volunteers developed PTSD [24]. In the first 10 to 34 months, individuals who were directly exposed showed a prevalence of PTSD evolving from 24% to 17%, while for those who were indirectly exposed it went from 4% to 2% [71]. In France, following the November 2015 Paris terrorist attacks, prevalence amongst resident physicians was 12.4% [16] and between 3.4-9.5% in other first responders [44].
Pre-attack risk factors to developing PTSD include being a woman, being of Asian or Hispanic decent (in the American context), having been exposed to a previous terror attack, experiencing a traumatic event in childhood or adulthood, having low social and educational status and having pre-existing psychiatric comorbidities [8,13,15,22,28,33,40]. One study found that a genetic polymorphism of the serotonin transporter (5-HTT (5-hydroxy tryptamine)] gene) may have led to higher rates of PTSD post 9/11 [45]. Personality characteristics associated with PTSD include negative affectivity, detachment and psychoticism, as well as less perceived self-efficacy [51,54]. In first responders, having only basic life-saving training versus more intermediate or advanced training, was found to be a risk factor for PTSD [31]. During the terrorist attack, the main predictors for developing PTSD are level of exposure [48,73], including experiencing high perceived threat and having witnessed a life-threatening injury [12,36]. Higher perceived threat is a predictor for developing PTSD even in individuals who did not directly witness the attacks [36]. Following the terrorist attack, having low social supports, comorbid depression, anxiety and alcohol use have been shown to be risk factors for developing PTSD [8,13]. Suffering a physical injury secondary to the terrorist attack, regardless of the severity of the injury, is one of the biggest predictors of developing severe PTSD [22].
Regarding first responders, having had only basic life-saving training, as opposed to intermediate or advanced training, as well as having to intervene on unsecured crime scenes, likely leading to higher fear of death, were found to be risk factors for developing PTSD [31,44]. Certain studies also commented on risk factors associated with increased severity of PTSD. These include low social integration into the community, higher level of exposure to the attack, job loss following the event, marital status, unmet mental health needs, low education and socio-economic status, being a female and being of Hispanic descent [27,42,52,60]. In regard to symptomatology and comorbidities, risk factors for more severe PTSD include having severe hyperarousal symptoms, experiencing bereavement, being injured by the attack, having a history of PTSD, depression or anxiety pre-attack, having other medical conditions diagnosed post-attack, higher levels of exposure to the attack and a lifetime trauma burden, especially post attack [27,42,60]. Finally, from a temperament perspective, using coping strategies such as substance use and avoidance, as well as callousness and perceptual dysregulation personality traits, can worsen the trajectory of the illness [51,60].

Protective Factors
When individuals and communities are exposed to terrorism, certain factors have been shown to protect against the development of mental illness. With regard to other forms of trauma, the general understanding is that adaptive coping strategies, greater social support and a sense of purpose are linked to lower PTSD symptoms [84]. These have similarly been shown to be protective factors in the context of terrorism exposure [60]. For first-line workers, feeling well prepared prior to the event, higher levels of training, feeling supported by leadership, lower role conflict, higher role clarity and predictability have shown to lead to lower rates of PTSD and less psychological distress [56,59,60]. More optimistic personality styles, benign styles of humour, perceived self-efficacy and the belief of having a life purpose all were traits that were associated with lower rates of psychological distress and post-traumatic symptomatology [54,55,57,60]. Individuals who employ problem solving and cognitive restructuring coping strategies were associated with fewer post-traumatic reactions and active coping skills distinguished between improving and chronic trajectories [60,62]. Finally, less severe emotional numbing symptoms was associated with higher rates of symptom recovery [27].

Symptom Clusters and Course of Illnesses
PTSD is often a chronic and highly disabling illness with an 18-50% recovery rate within the first 3-7 years. The four symptom clusters of PTSD include continuously reliving the traumatic event, persistent avoidance of stimuli related to the event, symptoms of emotional numbing and increased arousal response [84,85]. In first-line workers who intervened during terrorist attacks, especially volunteers, studies have shown that rates of PTSD continue to increase until a peak at 5-6 years post event. With regard to PTSD related to terrorism, even 6-7 years after the attack, 15-26% of direct victims continue to report PTSD symptoms [74]. Compared to other sources of PTSD, terrorism leads to a longer duration of illness (202 versus 92 months), with non-traditional workers showing the highest rates of chronic symptoms [67,75]. In one study, there was no statistically significant difference in the severity of symptoms between PTSD related to terrorism versus other forms of PTSD; however, higher avoidance symptoms were found, which is generally a severity marker [67]. However, one study out of Italy showed higher severity scores on the CAPS scale in terrorism than other forms of trauma [75]. When examining symptoms related to reliving the trauma, auditory reminders were the most frequently encountered and the most distressing [68,69]. One study found that the most central symptom seen in PTSD in the context of terrorism is feeling emotionally numb [66].

Major Depressive Disorder Prevalence and Risk Factors
There are less studies available that look into the impacts of terrorism on rates of MDD. Rates of new-onset MDD post 9/11 were 26% amongst individuals who were in the vicinity of the attacks, and 45% amongst those with a trauma-exposed close associate [21]. Rates of MDD amongst community members and rescue workers alike were found to be 15.3% 10-15 years post attack [19]. Prevalence of MDD 12 years post attack was found to be 17.2% for non-traditional rescue workers and 30.3% for police officers [10]. Another study suggested that the majority of individuals directly exposed to terrorism, do not suffer from MDD in isolation. Even 14-15 years after the attack, 6.8% had MDD alone while 8.9% had comorbid MDD and PTSD [54]. When looking at individuals meeting criteria for PTSD following a terrorist attack, 68.2% also had comorbid MDD [54]. Another study showed similar numbers, with a prevalence of MDD 14-15 years post-attack being 18.6%, and over half those cases being associated with a diagnosis of PTSD [5]. Once again, prevalence can vary when looking into communities around the world. In Tunisia, following the Bardo Museum terrorist attacks, 40.6% of museum employees endorsed depressive symptoms after 4-6 weeks [13]. Conversely, only 2.4% of resident physicians reported depressive symptoms after the 2015 terrorist attack in Paris [16].
There are a number of factors identified in the literature which may lead to certain individuals having a higher likelihood of developing MDD after a terrorist attack. In general, for those who were directly exposed or who lived in the vicinity of a terrorist attack, being less educated, of lower socio-economic status, unemployed and having lower social integration and support increase their risk [5]. Those factors, along with traumatic life events post terrorist attack and chronic physical illness decreased the likelihood of recovering from MDD [5]. In a Tunisian study, low social support seemed to be the best predictor for both PTSD and MDD symptoms in directly exposed individuals [13]. Regarding chronic symptoms of MDD in the American context, less social support as well as less perceived self-efficacy were risk factors [54].
Regarding the level of exposure, some evidence suggests that individuals who lost a loved one to a terrorist act are more than twice as likely as direct witnesses to develop MDD [21]. In Norway, parents of victims of the attack were three times more likely to develop MDD and anxiety than the general population [50]. Furthermore, parental emotional reaction worsened the more symptomatic their child became [38]. MDD appears to be related to the magnitude of the attack's impact on daily life, as well as how connected an individual is to the community affected [21]. In those who are not directly exposed to the attack, some evidence suggests that MDD can manifest only when an individual identifies the victims as being similar to their loved ones [37]. In those who were neither directly exposed nor had a loved one who was exposed, increased TV viewing related to the attack increased the likelihood of developing MDD, but only when it was associated with a decrease in perceived safety [32]. Similarly, in Nigeria, individuals who scored higher on the Terrorism Catastrophizing Scale were more likely to express symptoms of MDD and anxiety [7].

Anxiety Disorders Prevalence and Risk Factors
Individuals having survived a terrorist attack, especially those having suffered physical injuries, appear to report higher rates of anxiety disorders than the general population [25]. In the 10-11 years following the 9/11 attacks, 5.8% of police officers at the scene displayed comorbid anxiety and PTSD, and 47.7% had comorbid MDD and anxiety disorder [8]. Following the 13 November 2015 terrorist attacks which took place in Paris, 11.2% of resident physicians reported anxiety disorders [16]. In Denmark, rates of anxiety disorders saw a 16% increase following the Oslo bombings, and a 4% increase following 9/11 [17,18].
As with MDD, a multitude of factors can increase a person's chances of developing an anxiety disorder following a terrorist attack. Some factors, such as scoring high on the Terrorism Catastrophizing Scale or having a loved one who was exposed to the attack overlap with the risk of developing MDD [7,50]. In Europe following an ISIS truck attack, physical proximity to the event, ISIS anxiety and perception of danger increased the risk for psychological distress in general [41]. Not only is physical proximity an important factor, one study found that cultural proximity also led to increases in trauma and stressor related disorders following terrorism. After the Oslo attack, the population of Denmark saw a spike in these disorders, independent to media coverage, possibly because of cultural and geographic proximity to the victims [18]. Regarding symptom severity, the frequency of exposures to reminders of the attack can lead to a worsening of MDD and anxiety disorders, as well as a global decline in functioning [69]. Those who endured traumatic experiences in their adulthood, as well as recent life stressors, were at higher risk of worrying about future terrorist attacks [33].

Interventions
Prior to beginning this section, it is important to note that this paper is not a review of all treatment modalities. Comments on interventions are based off of the treatment modalities detailed in the articles sampled. Following a terrorist attack, one study showed that over 25% of individuals with PTSD or MDD had unmet mental healthcare needs over the last year [19]. Individuals who were more likely to seek out counselling were Caucasians, Hispanics, children at the time of the attack, those with higher levels of exposure, those who experienced peri-event panic attacks and those who had accessed counseling prior to the attack [79]. Individuals who were least likely to seek out mental health support after 9/11 were found to be African Americans, Asian Americans, those of lower educational or economic status, and those without a regular physician [79]. Similarly, a study out of the U.K. highlighted that 2/3 of individuals exposed to terrorism who were connected to mental health supports, did so via their family physician [76]. Of those who accessed mental healthcare, individuals who rated counseling to be helpful were likely female, African American, over the age of 65 or those with very high exposure [79]. A study following the Utoya massacre noted that early outreach programs provided benefit to exposed individuals with and without PTSD, MDD and anxiety disorders [78]. However, it highlighted challenges in reaching certain populations, specifically individuals in modern family structures and ethnic minorities [78]. It is important to consider new and innovative means of connecting both the population at large and hard to reach populations to services and supports following trauma. One such means that has been recently described in the literature is supportive text messaging [86]. By providing personalized support to patients, mobile phone technologies have been found to potentially improve the outcomes of a number of mental health conditions such as MDD and possibly PTSD [86][87][88][89][90][91][92]. Similarly, a study from the U.K. highlighted the benefits in a general screening program for PTSD, MDD, anxiety disorders and alcohol use following a terrorist attack [76].
Trauma-focused cognitive-behavioural therapy (T-f CBT) has been found to be the therapy of choice for PTSD in victims of terrorism [77]. According to the articles sampled in this review, there is less evidence available regarding treatments of other mental disorders in victims of terrorism, as well as in non-developed, non-Western countries [77]. One study from Madrid examined a sample of survivors of terrorist attacks that occurred an average of 23 years prior. After a course of t-f CBT, prevalence of PTSD went from 23% to 3.2%, and anxiety disorders went from 14% to 9.7% [80]. Following the course of t-f CBT, no participants were expressing symptoms of MDD or panic attacks [80]. Significant decreases in symptoms were still present at a 1-year follow-up [80].
Improving one's tendency to forgive and social supports, as well as working on other positive coping strategies, have been noted as possible therapeutic interventions following a terrorist attack, as these have been found to lower levels of PTSD symptoms [82]. Debriefing, and other crisis interventions in the first 24-72 h following terrorist attacks have been shown to lower quality of life and lead to worsening of MDD and PTSD symptoms [81,93]. According to the reviews sampled in this study, it was noted that there is limited information on treatment of mental illness following a terrorist attack. There is especially a lack of randomized-controlled trials in this area. However, there is some evidence highlighting the benefits of exposure-based approaches including virtual reality (VR) technology, as well as the use of SSRI medications [73].

Limitations
This literature review has some limitations. Firstly, the vast majority of articles studying the impacts of terrorism on mental illness occurred in developed countries, specifically the U.S.A; particularly in the post 9/11 context. This is the case even though 95% of deaths caused by terrorism occur in the Middle-East, Africa and South Asia [2]. Furthermore, articles in languages other than English were excluded, which further reduces the ability to assess studies published in other cultural contexts. Thus, it begs the question as to how culturally and globally representative the data presented in this review is.
Secondly, this article consists of a general literature search, not a systematic review or scoping review. The authors did not use the PRISMA or PICO rules and triangulation method in the search for articles. It was mainly completed by a single examiner though there was a second reviewer who aided in reviewing the admissibility of articles that did not fully meet the inclusion and exclusion criteria previously agreed upon. In this study, only one database (MEDLINE) was used, which can introduce a bias towards medical sources and overlook non-medical, like psychological, literature. This method of reviewing the literature can introduce selection bias. As a result, it is presented as a qualitative summary of evidence found in a variety of articles. Notwithstanding the limitations of this review, it still provides an insightful overview of prevalence, risk factors, protective factors, symptomatology and possible management option of a mental illness in the context of terrorism.
Lastly, it is important to note the lack of a widely agreed upon definition of terrorism. Indeed, the decision to deem one violent act as a terrorist attack versus another can often be subjective and may be racially biased. In general, there is a Western bias when defining events as terrorism or not, particularly in the post-9/11 climate. This bias is likely to have been reflected in the present study.

Conclusions and Future Directions
It is challenging to identify a globally representative value for the prevalence of mental illness following a terrorist act, as the populations studied have been quite heterogeneous. In general, it appears that directly and highly exposed individuals with chronic physical sequelae following the incident, and underprepared (often volunteers) first-line workers, are populations with higher rates of PTSD. Populations that are geographically and culturally close to the victim population, as well as loved ones of victims, seem to have higher rates of MDD and anxiety disorders than the general population. Given that screening and outreach programs have been found to be effective in providing consistent mental health support to at risk populations, we wonder if on a systems level this would be needed to reduce negative outcomes following terrorism.
As mentioned, there are many studies that examine the impacts of terrorism on mental disorders, specifically looking at prevalence and short-term illness progression. However, there are few studies that perform longitudinal assessments of illness trajectories in individuals exposed to terrorist acts. Studies of that nature would likely offer valid and valuable contributions to the field, especially with regard to the course of illness, and elucidating which factors and populations may be more at risk for developing chronic pictures versus resolution of symptoms. There would also be a benefit in further exploring the impact of terrorism on MDD and anxiety disorders, as there is at this point much more research looking into PTSD.
In the studies sampled, it has been noted that, with regard to the study of treatments and interventions post-terrorism, there is limited information, especially with regard to a lack of randomized-control trials. Though there have been some interventions that have been validated in the management of terrorism-related mental illness, such as t-f CBT and SSRIs, more information is needed, especially in terms of long-term outcomes.
Though this was not a topic specifically covered by this study, considering risk factors for becoming a terrorist can be extremely clinically valuable in reducing harm. Given the tremendous impacts on mental illness that terrorist acts have on the general population, developing ways to detect someone's risk of displaying terrorist behaviours is essential. Some papers have already compiled lists including social and clinical risk factors, and acts specific to terrorism that can be used as screening tools [94]. The UK's counterterrorism strategy, CONTEST, which was established in 2003 is one such team that is looking into this. The Terrorism Radicalization Assessment Protocol (TRAP-18) has been identified as a tool that may be helpful for clinicians. Further research looking into risk and protective factors for becoming a terrorist may be useful [94].
It is important to note once more that this paper is a qualitative, general literature search as opposed to a systematic review or a scoping review. As mentioned, this can introduce significant bias in the information presented. As a result, we feel that future studies implementing scoping review or systematic review methods would be extremely useful to mitigate this bias.