Two questions require attention when considering the implications of domestic violence for postdisaster recovery.
The first question is whether domestic violence increases in prevalence after disasters. There are only minimal data that are relevant to this question. Mechanic et al. (1) undertook the most comprehensive examination of intimate violence in the aftermath of a disaster after the 1993 Mid-western flood. A representative sample of 205 women who were either married or cohabitating with men and who were highly exposed to this disaster acknowledged considerable levels of domestic violence and abuse. Over the nine-month period after flood onset, 14% reported at least one act of physical aggression from their partners, 26% reported emotional abuse, 70% verbal abuse, and 86% partner anger. Whether these rates of physical aggression are greater than normal is not known because studies of domestic violence from previous years and under normal conditions have showed the existence of rates of violence as low as 1% and as high as 12%.
A few studies have produced evidence that supports the above. Police reports of domestic violence increased by 46% following the eruption of the Mt. St. Helens volcano (2). One year after Hurricane Hugo, marital stress was more prevalent among individuals who had been severely exposed to the hurricane (e.g., life threat, injury) than among individuals who had been less severely exposed or not exposed at all (3). Within six months after Hurricane Andrew, 22% of adult residents of the stricken area acknowledged having a new conflict with someone in their household (4). In a study of people directly exposed to the bombing of the Murrah Federal Building in Oklahoma City, 17% of noninjured persons and 42% of persons whose injuries required hospitalization reported troubled interpersonal relationships (5).
The second question is whether domestic violence, regardless of the reasons how or why it occurs, influences women's postdisaster recovery. An important finding from Mechanic et al.'s (2001) study was that the presence of domestic violence strongly influenced women's postdisaster mental health. Thirty-nine percent of women who experienced postflood partner abuse developed postflood PTSD compared to 17% of women who did not experience postflood abuse. Fifty-seven percent of women who experienced postflood partner abuse developed postflood major depression compared to 28% of nonabused women. Similarly, Norris and Uhl (3) found that as marital stress increased, so too did psychological symptoms such as depression and anxiety. Likewise, Norris et al. (4) found that 6 and 30 months after Hurricane Andrew, new conflicts and other socially disruptive events were among the strongest predictors of psychological symptoms.
These findings take on additional significance when it is remembered that not only are women generally at greater risk than men for developing postdisaster psychological problems, but women who are married or cohabitating with men may be at even greater risk than single women (6,7). In contrast, married status is often a protective factor for men (8,9). It also has been found that the severity of married women's symptoms increases with the severity of their husbands' distress, even after similarities in their exposure have been taken into account (7).
In summary, although the research regarding the interplay of disaster and domestic violence is not extensive and little of it has been derived from studies of incidents of mass violence, the available evidence does suggest that services related to domestic violence should be integrated into other mental-health services for disaster-stricken families. Screening for women's safety may be especially important. Helping men find appropriate ways to manage/direct their anger will benefit them and their wives. It will also help their children, as children are highly sensitive to postdisaster conflict and irritability in the family (7,10).