In the absence of baseline data in most studies, the literature on pre-disaster risk factors is rather limited. To date, the most reliable predictors of postdisaster psychopathology are female sex and especially being a mother of young children . After the TMI disaster, women with young children showed significantly increased rates of anxiety and depressive disorders compared with non-exposed controls, as assessed with the Schedule for Affective Disorders and Schizophrenia-Lifetime (SADS-L) (risk ratio 3.4 for new cases ). In a within-sample analysis of risk factors among a large population sample in Belarus exposed to the Chernobyl disaster, Havenaar et al.  found that being a mother was associated with a 4-5-fold risk of having a DSM-III-R anxiety disorder and an almost 3-fold risk of any psychiatric disorder. However, these variables are also risk factors for poor mental health in non-disaster studies [100-102]. Nevertheless, the consistency of the findings in disaster studies suggests that secondary prevention efforts addressing PTSD, depression and anxiety should target women, especially those with young children.
It is also important to note that the types of behaviors that are more likely to be seen in men, such as substance abuse, have only rarely been included in disaster studies. Thus, it remains to be seen whether the difference we currently observe is due to limitations in measurement (e.g., a circumscribed view of the phenotype!).
It is noteworthy that the elevated rates of psychopathology in women have also been found in some studies of children, such as some of the studies after Hurricane Andrew .
Several studies have shown that a personal history of psychopathology is a risk factor for poor mental health after a disaster [6, 47, 75, 100, 103]. The Bromet et al. research program on the mental health effects of the 1979 nuclear power plant accident at TMI was the first large-scale study to use a semi-structured diagnostic interview schedule. That study showed that pre-accident history of depression and anxiety disorders was among the most significant predictors of post-TMI depression and anxiety.
Long-term outcome has also been linked to mental health history. For example, Weisaeth  showed that outcome 4 years later was significantly influenced by pre-exposure psychological functioning. Similarly, McFarlane  found a significant association between a history of psychiatric disorder and chronic PTSD in a large group of firefighters in Australia assessed after a major bush fire. These findings are consistent with general population studies of PTSD showing that a personal history of psychopath-ology is an important risk factor [106-108]. However, findings in this area are not entirely consistent. In the only prospective study available, Robins et al.  did not find that after exposure to floods or to dioxin, people with a history of mental health problems had a higher than expected rate of new or recurrent episodes of psychiatric problems, either at the symptom level or at the level of clinical disorders. However, this study may have been "underpowered" and hence unable to detect this effect. Retrospective reporting bias could of course influence the findings in these studies, since the psychiatric history reports are retrospective in nature.
There is also growing evidence that disaster survivors who were exposed to traumatic life experiences before the event are more vulnerable to their impact . There is also evidence that children of survivors of severe traumatic events, such as the Holocaust, may be more at risk of developing PTSD and related disorders . While this area needs further investigation, it is consistent with Turner et al.'s  research showing that individuals exposed to early life trauma, continuing strain and acute stressors are at increased risk of adverse mental health outcomes.
In child survivors, it appears that the most important risk factor is the mothers' response to the disaster [48, 89, 91]. Several child disaster studies, including our TMI work on very young children  and school-age children , as well as the study by Laor et al.  of preschool children after the Scud missile attack on Tel Aviv (Israel), found that mothers' response was the more significant factor. However, both the age of the sample and the extent of involvement in the disaster can influence the contribution of the mothers' response. For example, in a study by Pynoos et al.  of elementary school children in Los Angeles exposed to a fatal sniper attack on their playground, proximity to the violence was the most important predictor of type and number of PTSD symptoms. In a subsequent study of the Armenian earthquake, Pynoos et al.  demonstrated a similar correlation between proximity to the epicenter of the quake and severity of children's post-traumatic stress reactions. March et al.  showed a clear dose-response effect, with the greatest PTSD symptoms reported by children who witnessed an industrial fire and had a relative or friend hurt or killed, followed by those with a relative or friend who was hurt or killed (only), and those who witnessed the event (only), with the lowest level among those with none of these exposures.
Was this article helpful?