The present review examines the relationship between disaster occurrence and psychopathology outcome for 52 studies that used quantitative measures of such a relationship. Descriptive and inferential techniques were used to examine relationships among four sets of variables: (a) the characteristics of the victim population, (b) the characteristics of the disaster, (c) study methodology, and (d) the type of psychopathology. A small but consistently positive relationship between disasters and psychopathology was found. The distribution of effect-size estimates was significantly heterogeneous, and this heterogeneity was partially accounted for by methodological characteristics of the research. When controlling for methodology, victim and disaster characteristics also contributed variance to the disaster-psychopathology relationship. Implications for future research are outlined in view of these results.