Evaluating The Subtypes of Panic Disorder by using Principal Component Analysis

Evaluating the subtypes of panic disorder by using principal component analysis Objective: Panic disorder is put forward to have a heterogeneous structure and multiple subtypes. Evaluating the different subtypes is important for diagnosis, treatment and prognosis. Studies about the quantitative psychopathology assessment are so rare. We aim to determine the subtypes with principal component analysis of panic attack symptoms. Method: 159 outpatients with panic disorder who had referred to the anxiety disorder outpatient unit in Bakirkoy Research and Training Hospital for Psychiatry, Neurology and Neurosurgery were included in the study. A sociodemographic questionnaire that was prepared by the investigators, The Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I), Panic and Agoraphobia Scale (PAS), Hamilton Depression Rating Scale (HAM-D), Beck Anxiety Inventory (BAI) were used in the study. Variables with 15 items were created adding ‘Fear of stroke’ and ‘desire to escape, restlessness’. Principal Component Analysis was applied according to the panic symptoms that the patiens declared. Results: Six loading factors were obtained explaning 56% of total variance in Principal Component Analysis. The first set was named autonomic activation, the second was vestibular symptoms, the third was cardiovascular symptoms, the fourth was pseudoneurologic symptoms, the fifth was respiratory system symptoms and the sixth was fear of death. Conclusion: Symptom profiles in our study were compatible with DSM-IV-TR diagnostic criteria. Significant panic disorder subtypes were found in principal component analysis.