This study will compare the effectiveness of family- and peer-oriented therapy in treating children with anxiety disorders and will also test for therapy specificity effects and potential mediators of outcome.

Data suggest that individual Cognitive Behavior Therapy (CBT) is effective in reducing anxiety disorders in children. The incorporation of parents and peers in a CBT treatment program also has been found to be effective in reducing anxiety symptoms, because a child's environment (i.e., parents, peers) affect the development, course, and outcome of childhood anxiety and functional status. Children and their parents will be randomly assigned to one of two treatment conditions: Family/Parents CBT (FCBT) or Peer/Group CBT (GCBT) for 14 to 16 weekly sessions. Parent-child relationships are the focus of FCBT. Parents will be trained to manage their children's anxiety and avoidant behaviors by increasing acceptance and warmth toward their children. Children in the GCBT group will be trained to be more helpful and positive toward other children through role-playing activities. Interviews, questionnaires, and behavior observation tasks will be used to assess participants. All participants will be assessed at pretreatment, post-treatment, and at yearly follow-up visits. The first set of hypotheses that will be tested is that FCBT will produce significantly greater specific effects on parenting skills and parent-child relationships than on child social skills and peer-child relationships. GCBT, on the other hand, will produce significantly greater specific effects on child social skills and peer-child relationships than on parenting skills and parent-child relationships. The second set of hypotheses will test whether or not it is the changes that are produced on these variables that mediate treatment response. Thus, the second set of hypotheses that will be tested is that parenting skills, parent-child relationships, child social skills and/or peer-child relationships will be significant mediators of treatment response, i.e., anxiety reduction

Inclusion criteria:

• DSM-IV criteria for Generalized Anxiety Disorder (GAD), Social Phobia (SP), or Separation Anxiety Disorder (SAD)
• Mean score >= 4 on the Clinician's Rating Scale of Severity
• Discontinuation of all other psychosocial treatment upon consultation with clinic staff and the service provider

Exclusion Criteria:

• DSM-IV criteria for a disorder other than GAD, SP, and SAD
• Diagnosis of any one of the following: pervasive developmental disorders, mental retardation, selective mutism, organic mental disorders, or schizophrenia and other psychotic disorders
• At risk for harm to self or others