The taijin kyofusho (TKS) offensive subtype is thought to be a culture-bound syndrome similar to social anxiety disorder (SAD). In Western countries, such patients would be diagnosed as having delusional disorder, somatic subtype, or body dysmorphic disorder. Recently, open trials for the TKS offensive subtype and a randomized controlled trial for body dysmorphic disorder demonstrated that selective serotonin reuptake inhibitors (SSRIs) might be as effective in TKS as in SAD. This study investigated the efficacy of the SSRI paroxetine in patients with the TKS offensive subtype, both on anxiety and fears, as well as insight. This study was a 12-week open trial using paroxetine in 22 patients with TKS. Subjects were diagnosed based on the original diagnostic criteria for the TKS offensive subtype. Insight regarding TKS symptoms was assessed by the 11th supplement subscale "Insight into obsessions and compulsions" of the Yale-Brown Obsessive Compulsive Scale (Y-BOCS). The Liebowitz Social Anxiety Scale (LSAS), the Beck Depression Inventory (BDI), the State-Trait Anxiety Inventory (STAI), Rosenberg's Self-Esteem Scale, and the Interpersonal Distrust subscale of the Eating Disorder Inventory were also administered. Offensive anxiety was assessed by the original TKS offensive anxiety subscale (0-3 points). The primary efficacy variable was the Clinical Global Impression scale (CGI) global improvement item. Nineteen patients completed the study. Forty-seven percent (9/19) were responders to the drug treatment (scoring 2 or less on the CGI). Last observation carried forward (LOCF) analysis (N=22) demonstrated a statistically significant reduction in LSAS total score and offensive anxiety in TKS, and the insight scale score of the Y-BOCS also significantly improved. Interpersonal distrust showed a trend toward improvement.