The most important predictors in our study were generally consistent with those found in previous post-disaster studies (Galea et al. 2005): prior psychopathology, disaster severity, and history of previous trauma. This adds support to the recommendation of North & Pfefferbaum (2013) to include information about these three classes of risk factors in needs assessment surveys of disaster survivors. It is also noteworthy that several previous epidemiological studies found, consistent with our result, that human-made disasters have more pernicious psychological effects than natural disasters (Galea et al. 2005), although this association became much less pronounced when we controlled for disaster-related characteristics, suggesting that at least part of the reason human-made disasters are associated with higher rates of PTSD than natural disasters is that the former are objectively more severe. Caution is needed in interpreting this result, though, as an exploratory factor analysis of TEs in an earlier WMH report found that the human-made disasters reported in the WMH surveys include a mix of accidents caused by human error and motivated acts of terrorism (Benjet et al. 2016). We have no way of distinguishing these two types of human-made disasters to determine if they have similar associations with PTSD.
