From the Journals

Scales assessing pediatric OCD possess little symptom overlap



There is considerable heterogeneity among symptoms included in various freely available, self-reported scales assessing obsessive-compulsive disorder in pediatric populations, according to Rachel Visontay and her associates at the University of New South Wales, Sydney.

Illustration featuring the letters OCD wildpixel/Thinkstock

They reviewed seven scales that exclusively assessed obsessive-compulsive disorder (OCD) and measured both obsessive and compulsive components in their study, published in the Journal of Obsessive-Compulsive and Related Disorders. The scales are:

  • Children’s Florida Obsessive Compulsive Inventory.
  • Children’s Yale-Brown Obsessive Compulsive Scale Symptom Checklist.
  • Leyton Obsessional Inventory–Child Version.
  • Children’s Obsessional Compulsive Inventory–Revised–Self Report.
  • Obsessive Compulsive Inventory–Child Version.
  • OCD Family Functioning Scale.
  • Short OCD Screener.

A total of 54 umbrella symptoms were included over all seven scales: 32 obsessions, 21 compulsion, and 1 other symptom (Give poor class presentations despite planning). Half of these symptoms were unique to one scale but were more commonly obsessions (18 of 32) than compulsions (8 of 21). No obsession symptom appeared on more than five of the seven scales, but two compulsion symptoms (cleaning and checking compulsion) did appear on all scales.

The mean overlap between scales after Jaccard analysis was 0.14 for obsessions and 0.39 for compulsions, indicating very weak and weak overlap, respectively. The correlation between number of included symptoms and mean overlap was 0.43 for obsessions, indicating that scale length did play a role in determining overlap, but was –0.09 for compulsions, indicating no relationship between scale length and symptom overlap.

“While youth OCD scales generally measure the same broader construct, they might have different scopes within that construct, or may be more or less comprehensive than one another. As such, low content overlap does not necessarily mean a scale is ‘worse’ than one with higher overlap,” the investigators wrote. “However, researchers and clinicians should be careful in selecting scales to assess OCD symptomatology, being mindful of the specific characteristics of each instrument.”

The study authors reported no conflicts of interest. The study was funded by a University of New South Wales Medicine, Neuroscience, Mental Health and Addiction Theme and SPHERE Mindgardens Clinical Academic Group collaborative research seed funding grant.

Next Article: