IDENTIFYING PANIC DISORDER SUBTYPES USING FACTOR MIXTURE MODELING.

Area of Science:

• Psychiatry
• Clinical Psychology
• Anxiety Disorders Research

Background:

• Panic disorder (PD) presents with high heterogeneity, challenging etiological, neurobiological, and treatment research.
• Previous attempts to identify PD subtypes lack independent reproducibility, leading to inconclusive literature.

Purpose of the Study:

• To identify and validate distinct subtypes of panic disorder using a data-driven approach.
• To characterize these subtypes based on a wide range of clinical features.

Main Methods:

• Employed factor mixture modeling, a case-centered approach, on anxiety symptom data from the Netherlands Study of Depression and Anxiety (NESDA) cohort (n=658).
• Utilized the Beck Anxiety Inventory to assess a broad spectrum of anxiety symptoms.
• Validated identified subtypes using diverse clinical characteristics.

Main Results:

• A three-class, one-factor model was the best fit, identifying a cognitive-autonomic subtype (29.8%), an autonomic subtype (29.9%), and an aspecific subtype (40.3%).
• Cognitive-autonomic and autonomic subtypes exhibited significant differences from the aspecific subtype in comorbidity and suicide attempt rates.
• Severity differences were noted between the cognitive-autonomic and autonomic subtypes, but not with the aspecific subtype.

Conclusions:

• Identified three qualitatively distinct PD subtypes: severe cognitive-autonomic, moderate autonomic, and mild aspecific.
• Subtype differences correlate with variations in severity, comorbidity, suicide attempts, sleep disturbances, and sense of mastery.