Predictors of complicated grief

Abstract This study examined whether process variables predict an outcome of complicated grief. A turbulent and prolonged grief was predicted to occur after the death of a spouse in subjects who had self-blame, used the deceased for an extension of self, had ambivalence toward the deceased, or overcontrolled emotional responses. Ninety subjects were examined at 6, 14, and 25 months after the loss via a self-report battery of process variables and a structured clinical interview designed to assess symptoms. Subjects with complicated grief were compared with those with normative grief. Predictions were supported only in tertiary data analyses; they were not supported well in the primary and secondary statistical analyses. The authors concluded that either self-reports of process variables are inadequate measures or the theory that led to these measures and predictions is in need of revision.