Parent-Child Diagnostic Agreement on Anxiety Symptoms with a Structured Diagnostic Interview for Mental Disorders in Children

Lukka Popp, Murielle Neuschwander, Sandra Mannstadt, Tina In-Albon, Silvia Schneider

Objective: In clinical structured diagnostic interviews, diagnoses based on parent and child reports have low to moderate agreement. The aims of the present study are (1) to examine diagnostic agreement on anxiety disorders between parents and children on the levels of current and lifetime diagnostic category and diagnoses focusing in particular on diagnostic criteria and (2) to identify parent- and child-related predictors for diagnostic agreement. Method: The sample consisted of 166 parent-child dyads interviewed with the Structured Diagnostic Interview for Mental Disorders in Children (Kinder-DIPS, Schneider et al., 2009). The children (51.8% girls) were between the ages of 7 and 18 years (M = 10.94; SD = 2.22). Results: Overall, parent-child agreement on the diagnostic category of anxiety disorder (k = 0.21; k = 0.22) and the specific anxiety diagnoses (base rate > 10%) of social phobia, specific phobia and separation anxiety disorder (k = 0.24-0.52; k = 0.19-0.43) and corresponding diagnostic criteria (k = 0.22-0.67; k = 0.24-0.41) were low to moderate with the highest agreement on separation anxiety disorder (k > 0.43). Lower maternal depression, and higher social support reported by mother and father were associated with higher parent-child agreement. Maternal depression was indicated as the strongest predictor. Parental sense of competence, parental anxiety, the amount of parent-child interaction and the child's age and gender had no predictive value. Conclusions: Parent-child agreement can be expected to be higher on the level of anxiety criteria compared to specific anxiety diagnoses and diagnostic anxiety category. Psychological strains in the family-especially maternal depression and low social support-lower the parent-child agreement on anxiety symptoms. Child- and relation-related variables (age, gender, amount of time parent(s) and children interact) play no role in the prediction of low parent-child agreement.