Empirical testing of an algorithm for defining somatization in children.

Area of Science:

• Pediatric Psychology
• Child Psychiatry
• Behavioral Science

Background:

• A prior algorithm proposed classifying children's somatization into well, medically ill, and somatizer categories.
• Empirical validation was suggested to confirm the algorithm's utility in clinical settings.

Purpose of the Study:

• To empirically validate the Postilnik et al. (2006) somatization classification algorithm.
• To assess the Child Behavior Checklist's (CBCL) effectiveness in differentiating these categories.

Main Methods:

• Parents of 126 children in pediatric clinics completed the CBCL.
• Physicians provided additional data via questionnaires.
• Statistical analyses included ANOVA and discriminant function analysis on CBCL subscales.

Main Results:

• Significant differences were found across categories on multiple CBCL scales, including somatic complaints, social problems, and internalizing behaviors.
• Discriminant function analysis achieved high accuracy for somatizers (78%) and well children (66%).
• Classification accuracy for medically ill children was lower (35%).

Conclusions:

• The somatization classification algorithm demonstrates potential for classifying children and adolescents with somatic symptoms.
• The CBCL appears useful in supporting this classification, particularly for identifying somatizers.