The rates of PTSD are very dependent on the sampling used in a study as well as the severity and nature of the event. Therefore, each new disaster should be considered as a novel event and predictions about the rates of morbidity should depend on careful consideration of which group of victims are being considered as well as the time that has elapsed following the disaster.
The Buffalo Creek disaster (dam break), which occurred in 1972, is one of the best studied disasters: a 59% PTSD lifetime rate was found among the victims, with 25% still meeting PTSD criteria some 14 years after the event . One of the highest rates was demonstrated by Goenjian et al.  following the Armenian earthquake, with 67% meeting PTSD criteria 18 months after the earthquake. In a study conducted 1 to 4 months after Hurricane Andrew , 33% met the criteria for PTSD. However, in some studies of low-exposure groups, the rates of PTSD are sometimes little different from the prevalence in the general population . Shore et al.  examined the impact of the Mount St. Helens volcanic eruption and compared the exposed population with a control group: the lifetime prevalence of PTSD in the Mount St. Helens group was 3.6% compared to 2.6% in the control.
In a study of an earthquake in Yunnan, China, three villages at increasing distance from the epicentre were evaluated . The rate of PTSD was 23.4% in a village where most houses were destroyed compared with 16.2% where only minor damage occurred.
Gender is also an issue influencing the prevalence, with 20% of men and 36% of women suffering from PTSD one year after a mass shooting . The lasting impact of events on children was demonstrated by the findings of a 33-year follow-up of victims of the Aberfan disaster, where rates of 29% current and 46% lifetime PTSD were found .
Similar issues influence the rates of other disorders following disasters. These disorders may emerge as comorbid conditions with PTSD, in which case they are likely to represent complications of PTSD and an indication of the severity of the underlying traumatic stress response. One review suggests that PTSD occurs four times more frequently in conjunction with comorbid diagnoses than it does alone, even in close proximity to the event .
Alcohol usage is often a response to the development of symptoms in disaster-affected populations, an issue which has emerged particularly amongst emergency service workers . On the other hand, there is an increasing body of evidence that depressive and anxiety disorders may emerge following traumatic events in the absence of PTSD . For example, there is an intuitive rationale in the potential role of loss in the onset of depression and of threat and horror as determinants of anxiety disorders . While there have been some explorations of the role that different types of disaster experiences play, in association with risk factors, in the onset of these disorders, this remains an area which needs further investigation . On the other hand, there are studies which have not found an increased prevalence of these non-PTSD disorders, despite high rates of PTSD .
One of the problems in defining the prevalence of these disorders is that their assessment requires in-depth structured diagnostic interviews, whose application is time-consuming and sometimes difficult in disaster victims. Furthermore, reported rates of anxiety and depression are often derived from continuous scales, making the clinical and diagnostic interpretation of these data difficult. A further methodological issue is that defining the onset of major depressive disorder or panic disorder is more difficult than for PTSD, which can be tied to a specific event by the content of the intrusive memories. Depressive and anxiety disorders are common in community samples in the absence of disasters, so that the accurate attribution of their onset to a disaster is a difficult task.
The available evidence suggests that depression is the second most common disorder to emerge in the aftermath of a disaster . One important issue is how depression following a disaster relates to preexisting morbidity in the community. Bravo et al.  studied the impact of a mud slide and flood in Puerto Rico which killed 800 people. Fortuitously, a year before they had studied this population and were able to re-evaluate 375 of their initial subjects. They found a significant increase in the symptoms of depression and a range of somatic complaints from the pre-disaster levels but failed to demonstrate any increase in panic disorder or alcohol abuse. An increased prevalence of PTSD was identified. Smith et al.  investigated a series of disaster events in St. Louis involving exposure to dioxin, floods and tornadoes. Exposed individuals had high levels of new PTSD symptoms. However, depressive symptoms increased only in those who had had previous depression.
Following the 1988 Armenian earthquake, a high-impact disaster, major depressive disorder was found in 52% of a stratified sample of 1,785 individuals. Depression was the only disorder in 177 of these individuals and was particularly associated with exposure and loss .
In summary, the rates of disorders such as depression are highly variable and affected by the intensity and nature of the disaster. Estimates of the health service needs of a population should take account of the nature of the event and cannot be easily derived from the literature. The question of the rates of anxiety and depressive disorders in the aftermath of disasters is an important issue for further investigation and must consider the interaction with the existing morbidity in the community. The physical presentation of psychiatric disorders also requires better clarification, since the somatic expression of distress in the aftermath of disasters has major practical implications for the post-disaster health services [65,69].
Was this article helpful?