Psychometric properties of the mini-social phobia inventory

Although a potentially useful measure, to date, there has been only one published test of the psychometric properties of the Mini-Social Phobia Inventory (Mini-SPIN). Therefore, the psychometric properties of the Mini-SPIN, a brief 3-item screen for social anxiety disorder, were examined.

Participants were 186 patients diagnosed with social anxiety disorder (DSM-IV criteria) attending a specialized anxiety disorders clinic for treatment, and 56 nonclinical participants were recruited to serve as comparisons. Participants were diagnosed using the Anxiety Disorders Interview Schedule for DSM-IV, and they also completed the Mini-SPIN, the Social Interaction Anxiety Scale (SIAS), and the Social Phobia Scale (SPS). Construct validity for the Mini-SPIN was assessed by its correlations with the SIAS and the SPS. Reliability, internal consistency, discriminant validity, and sensitivity to change were also examined, and receiver operating characteristic curve analysis was conducted to determine guidelines regarding cutoff scores for the Mini-SPIN. The study was conducted between April 1999 and December 2001.

Supporting findings from a previous study, strong support was found for the Mini-SPIN's ability to discriminate individuals with social anxiety disorder from those without the disorder. Receiver operating characteristic analysis revealed that using a cutoff score of 6 or greater (P < .001), the Mini-SPIN demonstrates excellent sensitivity, specificity, and positive and negative predictive values.

Findings suggest that the Mini-SPIN is a reliable and valid instrument for screening social anxiety disorder in adults. Importantly, the use of the Mini-SPIN in primary care may be one way to address the underrecognition of social anxiety disorder in such settings. Due to the ease and brevity of the measure, it also shows potential for use in epidemiology. Given that this study has revealed the ability of the Mini-SPIN to reflect treatment change, the Mini-SPIN may also be considered for use in treatment outcome studies that specifically require minimal assessment.