The purpose of this study is twofold: (1) to compare the effectiveness of two promising treatments for social phobia, a new cognitive therapy model (Clark & Wells, 1995; Borge et al., 2001) and interpersonal therapy (Lipsitz, Markowitz, & Cherry, 1997), adapted for inpatient groups; and (2) to study the empirical change processes in these therapies and compare them with the cognitive and interpersonal models of change.

Social phobia typically leads to severe impairment in work and other social life, and - without treatment - it can persist for decades. Given its prevalence, severity, and chronicity, effective treatment methods are strongly needed. However, traditional psychological and pharmacological treatments have had statistically significant, but clinically limited effects (Taylor, 1996).

Based on an empirical analysis of the cognitive processes in social phobia, Clark and Wells (1995) have developed a new cognitive model. Cognitive therapy (CT) derived from this model has been found to be superior to a combination of fluoxetine and self-exposure (Clark et al., 2003).

Social phobic symptoms may be viewed as a result of more general interpersonal difficulties and interpersonal psychotherapy (IPT) can be a reasonable alternative for social phobia (Lipsitz & Markowitz, 1996). A clinical case series indicates that social phobic patients improve during interpersonal psychotherapy for social phobia (IPT-SP; Lipsitz et al., 1999.

Inclusion Criteria:

• DSM-IV criteria for social phobia
• the patient consider social phobia as the main current problem
• willingness to suspend use of psychotropic medication, alcohol and other substances
• acceptance of random allocation
• ability to speak Norwegian
• age 18-65 years.

Exclusion Criteria:

• a history of recurrent major depression currently treated sucessfully with antidepressant medications
• has immediate need for additional treatment
• current psychotic disorder or substance abuse
• organic mental disorder
• previously treated with similar models