Social anxiety and ideas/delusions of reference in early psychosis

Abstract

Background: Social anxiety disorder (SAD) is increasingly recognized as a comorbid condition in early psychosis. Putative explanations include SAD as a psychological response emerging during recovery, a component per se or vulnerability marker for psychosis. Alternatively, it could also be an epiphenomenon of conceptually related symptoms, such as subthreshold suspiciousness and ideas/delusions of reference (I/DOR). This study describes the relationships between SAD and psychotic symptoms in stable patients.

Methods: A consecutive sample of 137 outpatients with early psychosis was interviewed using the Liebowitz Social Anxiety Scale (LSAS) and an I/DOR scale validated in Hong Kong Chinese. General symptoms were rated by patients' attending psychiatrists using the Positive and Negative Syndrome Scale (PANSS), Scale for the Assessment of Positive Symptoms (SAPS), Scale for the Assessment of Negative Symptoms (SANS), Calgary Depression Scale for Schizophrenia and Scale to Assess Unawareness of Mental Disorder. SAD was defined using LSAS cutoff as previously described.

Results: SAD was found in 45.3% of outpatients. Compared with patients in remission (defined using PANSS criteria, n = 106), SAD is more prevalent (39.6% vs. 64.5%) and severe (LSAS 28.7 vs. 44.1, both p = 0.01) in symptomatic patients (n = 31). LSAS correlated with all dimensions of the I/DOR scale. Patients with existing I/DOR are more likely to be socially anxious (OR 3.8, p = 0.00) and scored higher on LSAS (mean, 42.2) than those who no longer (28.3) or have never experienced I/DOR (26.9). Specific positive symptoms associated with social anxiety are PANSS suspiciousness (r = 0.28) and SAPS I/DOR (r = 0.31, both p < 0.01) but not other delusions or hallucinations. Insight is not associated with SAD.

Conclusions: SAD is more common and severe in symptomatic than remitted outpatients. It covariates with I/DOR, and is not affected by insight. Results suggest that SAD could be understood as closely related to I/DOR. Whether this relationship is correlational or epiphenomenological will require further exploration.