Prevalence and predictors of postdisaster major depression: Convergence of evidence from 11 disaster studies using consistent methods

Abstract

The objective of this study was to examine predictors of postdisaster major depression in two separate datasets of survivors of various disasters. Postdisaster major depression was examined in two disaster databases using consistent research methodology, permitting combination of databases into a combined dataset including 1181 survivors of 11 disasters representing all major disaster typologies with full diagnostic assessment using structured diagnostic interviews from two databases. The first database includes 808 directly-exposed survivors of 10 disasters. The second includes 373 survivors of the September 11, 2001 attacks on New York City's World Trade Center, recruited from employees of eight organizations affected by the disaster. This rich dataset permitted comparison of predictors of postdisaster major depression between databases and across survivors of different disasters. Identical models applied to both databases found postdisaster major depression to be independently associated with pre-existing major depression, indirect exposure to disaster trauma through family/friends, and disaster-related PTSD. In a final model limited to directly-exposed disaster across both databases, postdisaster major depression was independently associated with terrorism in addition to the 3 variables that predicted postdisaster major depression in the two separate databases. Replication of findings from one model to the next across different types of disasters and populations in this study suggests that these three variables could potentially provide a powerful tool for estimating likelihood of postdisaster major depression.

Keywords: Depression; Disaster; Predictor; Trauma.

Figure 1

Panel (a). ROC curves for predictors of postdisaster major depression: 10-disaster data 3-variable model. Panel (b). ROC curves for predictors of postdisaster major depression: 9/11 WTC data 3-variable model. Panel (c). ROC curves applying predicted probabilities of postdisaster major depression in the 10-disaster 3-variable model to the 9/11 WTC 3-variable model across all predicted probability cut points. Panel (d). ROC curves applying predicted probabilities of postdisaster major depression in the 9/11 WTC 3-variable model to the 10-disaster 3-variable model across all predicted probability cut points.