Item Response Theory Analysis: PROMIS® Anxiety Form and Generalized Anxiety Disorder Scale

Area of Science:

• Psychological and Cognitive Sciences
• Clinical Psychology
• Psychometrics

Background:

• Limited research exists comparing the psychometric properties of the PROMIS anxiety short form 6a and the GAD-7 scale.
• Item response theory (IRT) models provide a robust framework for evaluating the performance of patient-reported outcome measures.
• Anxiety assessment in veteran populations is crucial for effective mental health interventions.

Purpose of the Study:

• To compare the item functioning of the PROMIS anxiety short form 6a and the GAD-7 scale using IRT models.
• To evaluate the psychometric performance of both measures in a sample of at-risk U.S. military veterans.
• To determine potential advantages of one measure over the other for assessing anxiety in this population.

Main Methods:

• Secondary analysis of self-reported anxiety assessments from 67 at-risk U.S. military veterans.
• Application of item response theory models to compare item functioning.
• Evaluation of model fit, item discrimination, and item/test information curves.

Main Results:

• Both the PROMIS anxiety short form 6a and the GAD-7 scale demonstrated adequate model fit and acceptable item discrimination.
• Item and test information curves for both measures were unimodal and symmetric.
• The PROMIS anxiety short form 6a exhibited a wider and more even distribution of item difficulty estimates and less floor effect in response categories compared to the GAD-7 scale.

Conclusions:

• Both measures are potentially appropriate for assessing anxiety in veterans.
• The PROMIS anxiety short form 6a may be preferable, particularly for individuals with low-to-moderate anxiety, due to its item characteristics.
• Further validation in larger and more diverse veteran samples is recommended.