It is widely believed that personality disorders and/or traits (PDT) have significant impact on the phenomenology, severity and consequently the treatment of anxiety disorders (AD). Specific PDT's are thought to be characteristic of certain types of AD's. However, little experimental data support these assumptions. The interpretation of the few comorbidity and outcome studies investigating the role of PDT's in anxiety is problematic from both theoretical and methodological points of view. The authors review what is known about the co-occurrence of PDT's and some of the AD's. Particular attention is paid to studies that demonstrate the alteration of PDT's as a result of successful treatment of an AD and to those assessing the difficulties encountered in the treatment of AD's in the presence of certain PDT's. Specific recommendations for much needed research are given.