Introduction: Child’s insecure attachment to parents and insecure parental attachment has been linked to childhood anxiety (Brumariu & Kerns, 2010; Manassis et al.,1994).Whether attachment patterns can predict treatment outcome, is yet to be investigated. We examined the role of children’s attachment to parents, and parental attachment in predicting treatment outcome in anxious children receiving cognitive-behavioral treatment. Method: A total of 69 children aged 7-13 years were diagnosed at intake and post-treatment, using Anxiety Disorders Interview Schedule for DSM-IV (Silverman and Albano, 1996). In addition, children’s attachment to both parents was assessed using the Friends and Family Interview (Steele, Steele & Kriss, 2009) and Security Scale Revised (Kerns, Aspelmeier, Gentzler, & Grabill, 2001). Parents filled in the Experiences in Close Relationships-Revised (Fraley, Waller, & Brennan, 2000). We compared responders and non-responders with regards to attachment measures at intake to identify possible predictors. We then investigated the role of attachment in predicting treatment outcome in anxious children. Results: We did not find significant differences between attachment style in responders and non-responders in the present sample. We found a significant difference in maternal attachment anxiety scale (p=.011), with mothers of non-responders showing significantly higher attachment anxiety. Binominal logistic regression analysis was used to measure a predictive value of maternal attachment anxiety and it was found to significantly add to the model (Exp(B) =.958, CI0.95= [ .925, .994]; p=.021), even while controlling for symptom severity. Discussion: Maternal attachment anxiety was found to have a significant role in predicting treatment outcome. These results suggest that it is important to consider the parental attachment insecurity as a possible outcome hindering factor.