Background: The proposed draft of the DSM-5 from the Anxiety Disorder Workgroup recommends allowing the diagnosis of social anxiety disorder (SAD) in individuals with medical conditions, if the anxiety is considered to be excessive. Although prior research has examined diagnosing SAD in individuals with stuttering, such research has not yet been conducted in obese individuals. Methods: This study compared demographic and clinical characteristics of obese individuals diagnosed with DSM-IV SAD (n = 135), modified SAD (clinically significant social anxiety related to weight only; n = 40), and a group of obese individuals with no history of psychiatric disorders (n = 616). All participants were seeking psychiatric clearance for bariatric surgery and completed a comprehensive diagnostic interview. Results: The two social anxiety groups differed from the no disorder group on adolescent and past 5 years social functioning, and overall current functioning. Individuals with modified SAD had a later onset of their social anxiety, yet reported greater impairment in social life and distress about their social anxiety compared to the DSM-IV SAD group. Conclusions: Although both of the social anxiety groups differed from the no disorder group on social and overall functioning, there were few differences between those with DSM-IV SAD and modified SAD. This suggests that obese individuals with social anxiety related to weight only may experience comparable severity of anxiety to those with DSM-IV SAD, and supports adoption of the DSM-5 Workgroup's recommendation to change criterion H. Depression and Anxiety, 2011. © 2011 Wiley-Liss, Inc.