Approximately 20 000 people lost loved ones in the Great East Japan Earthquake on March 11, 2011. Death caused by disaster is sudden and unexpected, and sometimes includes additional trauma, such as facing life-threatening situations and witnessing damaged corpses. Raphael1 called these bereavements “traumatic loss,” which is more stressful, complicated, and difficult to recover from than the bereavement of natural death. Under traumatic loss, death by homicide, accident, and suicide is called “violent death.” This means death by intentional use of physical force or power, threatened or actual, against oneself, another person, or a group or community.2 Rynearson3 claimed that violent death comprised three Vs—violence, violation, and volition—and that these interfere with acceptance of death by the bereaved. Some studies have indicated that the prevalence of mental disorders, such as post-traumatic stress disorder (PTSD), complicated grief (CG), and depression among those who have experienced “violent death” was relatively higher than natural death.4-10 Dyregrov et al9 reported that the prevalence of PTSD (51% to 52%) and CG (78%) among survivors of suicide and accident. Violent death was associated with greater rates of PTSD and CG than those experienced by sudden infant death syndrome (PTSD 34%, CG 57%).9 In addition, the various negative emotions and cognitions, such as reprisal and guilt feelings and the socioenvironmental factors, including social reaction and stigma, that are associated with the aftermath of violent death, are considered to interfere with appropriate coping and contribute to persistent symptomatology. In terms of treatment, clinicians should consider interaction of these mental disorders and related issues. We focus on the characteristics of CG following violent death and the effect of PTSD on the complexity of grief symptoms.

Why is the prevalence of complicated grief after violent higher than that after other types of death?

In the general population, the prevalence of CG in those who have experienced loss of significant others has been reported as 2.4%11 to 6.7%,12 which is relatively low, but prevalence is higher among those bereaved by violent death (Table I).9,13-16 One reason for this is that violent death is sudden and unexpected. Suddenness and lack of readiness for death were reported as predictors of CG among the general population.11,17 Barry et al18 indicated that a lack of perceived preparedness for death was associated with severity of CG.

Prevalence of complicated grief among those bereaved by violent or nonviolent death. CBI, Core bereavement items; PG-13, Structured interview for Prolonged Grief Disorder; ICG, Inventory of Complicated Grief; SI-TG, Structured interview for traumatic grief; German ICG, German version of the Inventory of Complicated Grief Revised

Cause of deathAuthor (year)NPrevalence (%)Time since death (years)Measurement
Violent death
Bosnian conflictMomartin et al (2004)131263105CBI
Rwandan genocideSchaal et al (2009)144012.510PG-13
September 11th attacksNeria et al (2007)1570443.22.5-3.5ICG
Homicide and traffic accidentNakajima et aI (2009)167421.97.8SI-TG
SuicideDyregrov et al (2003)912878.01.3ICG
AccidentDyregrov et al (2003)96878.01.2ICG
Nonviolent death
CancerKersting et aI (2011)1221610.19.8German ICG-R
DiseaseKersting et aI (2011)121424.99.8German ICG-R

Violent death is not only sudden, but, importantly, is caused by violence, and it is significantly different from natural death in terms of the way it is thought about by the bereaved family. Currier et al10 reported that violent death made “sense-making” difficult. “Sense-making” is considered as an intermediate factor in that it is considered to help a bereaved family to accept death as a part of life. It is also responsible for the degree of severity of CG.10 The negative cognitive appraisal for themselves, others, and the world was another important mediating factor between violent death and following mental disorders such as PTSD, depression, and CG.16,19 In cases of violent death, the bereaved family is often exposed to the curiosity of the media and people around them, or may be slandered. Such experiences may affect the grieving process, as they could result in societal distrust, making it difficult for the bereaved to seek support, resulting in their social isolation. In fact, it was reported that bereaved families with mental disorders had a strong perception of being hurt by others after the death.16,19 In addition, in the case of violent death caused by crime, the influences of the legal process cannot be ignored.20,21 Legal proceedings such as police or attorney interviews and testimony in court might provoke psychological distress and PTSD symptoms by facing offenders, remembering details of the crime, and blame put on victims by defense attorneys.20,22,23 The outcome of the trial also affects mental health; with regard to the families of a homicide or traffic crime victim; it has been reported that their low satisfaction with the criminal justice system was associated with severity of PTSD, depression, and anxiety.4,24

The effect of post-traumatic stress disorder on complicated grief

Numerous studies have reported that a variety of mental disorders, such as depression, PTSD, and other anxiety disorders, coexist in bereaved individuals with CG.25-27 Simon et al26 indicated that 75.2% of patients with CG had at least ne axis I disorder of DSM-IV. Major depressive disorder and PTSD were prevalent comorbid disorders. In those bereaved by violent death with CG, prevalence of PTSD was reported to be as high as about 43% to 65 %13,15,28 (Table II). In circumstances of violent death, the bereaved frequently experienced life-threatening incidents or witnessed terrible scenes.28 Such traumatic experience is considered to contribute to the increasing prevalence of PTSD among those bereaved by violent death. Some studies ported that the severity of CG and PTSD was significantly positively correlated.12,14,19,25,29 It has been suggested that these conditions affected one other. In particular, intrusive symptoms of PTSD were associated with CG symptoms.13 It was indicated that intrusion was the comlon symptoms of both PTSD and CG.30-32 Findings from functional brain imaging also suggest the effect of PTSD on CG. It was reported that the amygdala, which was responsible for processing fear and anxiety, had exaggerated responses to general negative stimuli in PTSD.33 Furthermore, less activation of medial prefrontal cortex (mPFC), anterior cingulate cortex (ACC), and thalamus in PTSD subjects than non-PTSD subjects during fear activation was reported in previous studies.34,35 It was suggested that PTSD patients might have dysfunction of ACC and mPFC which played a role in suppressing excessive activity of the amygdala.34 There were a few studies on brain function with grief. Subjects with acute grief, a condition close to CG, also indicated that intrusion accompanied by strong sadness elevated the activity of the ventral amygdala.36 Therefore, the amygdala is responsible not only for feelings of fear, but also for separation distress. However, in contrast with PTSD, along with the elevated activity of the amygdala, the activity of the right ACC (rACC) was aIso elevated in grief subjects.36 This study indicated that le functional connectivity of the amygdala and the rACC had a negative correlation with the degree of sadness.36 The ACC and PFC play a role in the emotion regulation using cognitive reappraisal strategy.29,37 It is therefore assumed that activation of the rACC at the acute stage of grief contributes to the promotion of the normal grieving process. It is thought that the low activation of ACC at the early stage of grief in bereaved with PTSD leads to dysfunction of emotion regulation, resulting in interference with the normal grief process and developing CG.

Comorbidity of post-traumatic stress disorder (PTSD) and major depressive disorder (MDD) with complicated grief (CG).

Rateofcomorbidity (%)
Cause of deathAuthor (year)NPTSDMDD
Kosovar civilian warMorina et al (2010)282365.265.2
Not specifiedSimon et al (2007)2620648.555.3
(help-seeking CG patient)
September 11th attacksNeria et al (2007)1530443.336.0

It was reported the activation of nucleus accumbens, related to the reward system, was associated with CG, which was correlated with strong yearning for the deceased without being able to accept the death.38 Similarly, bereavement with PTSD is considered to be more difficult to accept the death than those without PTSD, because not only sadness, but also fear, might be evoked when recalling the deceased. In fact, it has been reported that PTSD, or its intrusion symptoms, was responsible for the severity of CG.9,13,39,40 Those reactions work to disrupt the normal grief process and contribute to the onset of CG.

The effectiveness of cognitive behavioral therapy for CG, including exposure to death, serves as evidence for the effect of PTSD on CG.31,41-43 Asukai et al43 modified the CG therapy31 for those bereaved by violent death, to focus more on an exposure exercises in traumatic situations, and reported that this modified treatment was effective for both symptoms of PTSD and CG. This result suggested that improvement of PTSD symptoms might act on reducing CG symptoms.


Violent death is not only sudden and unexpected, but threatens others by intentional power, resulting in significant impact on the mental health of bereaved persons. It was reported that there was 12.5% to 78% prevalence of CG9,13-16 among those bereaved by violent death. The factors affecting such high prevalence of CG following violent death are lack of readiness for the death, difficulty in sense-making, a high level of negative appraisal about the self and others, and various social stressors, such as exposure to the mass media, social stigma, and legal procedures. The comorbidity of PTSD was particularly considered to contribute to the development of CG by suppressing the functioning of the mPFC and the ACC, which facilitates the mourning process when grief distress is activated and interrupts acceptance of death. The DSM-5 working group is currently discussing whether CG as a bereavement-related disorder will be included in axis I mental disorders. However, its symptomatology and the biological basis of its pathology are unclear. It will be helpful to clarify the effect of PTSD on CG among survivors of violent death for understanding the pathogenic mechanism of CG and developing preventive intervention and treatment of CG.