To determine whether perinatal complications predict childhood anxiety disorders independently of parental psychopathology, we systematically assessed pregnancy and delivery complications and psychopathology in a sample of children (mean age=6.8 years) at high risk for anxiety disorders whose parents had panic disorder with (n=138) or without (n=26) major depression, and in contrast groups of offspring of parents with major depression alone (n=47), or no mood or anxiety disorders (n=95; total N=306). Psychopathology in the children was assessed by structured diagnostic interviews (K-SADS), and pregnancy and delivery complications were assessed using the developmental history module of the DICA-P. Number of pregnancy complications predicted multiple childhood anxiety disorders independently of parental diagnosis (odds ratio=1.6 [1.4-2.0]). This effect was accounted for by heavy bleeding requiring bed-rest, hypertension, illness requiring medical attention, and serious family problems. Associations remained significant when lifetime child mood and disruptive behavior disorders were covaried. Results suggest that prenatal stressors may increase a child's risk for anxiety disorders beyond the risk conferred by parental psychopathology alone.