Recent cognitive-behavioural models of obsessive-compulsive disorder (OCD) developed by Salkovskis (1985Salkovskis ( , 1989Salkovskis ( , 1998) ) and Rachman (1997Rachman ( , 1998) ) have advanced our understanding of the disorder substantially. Inherent in both models is the assumption that beliefs about obsessions are critical in the development and persistence of the disorder. Given that OCD is characterized by an aversive reaction to specific mental events, this emphasis on the cognitive processes involved in OCD is warranted. Wells (Wells and Mathews, 1994;Wells, 1997) proposed that our understanding of obsessional problems would be further advanced by examining the role of `metacognitive' processes and beliefs (that is, attempts to regulate thoughts and beliefs about thoughts and thought processes) in their development and persistence. In identifying metacognitive factors as unique contributors to the problem, this model offers new directions for treatment. This paper discusses the role of metacognition in obsessional problems in light of current models of OCD, and offers an elaboration of the role of metacognitive factors in the escalation and persistence of obsessional problems.