HTS: Liebowitz Social Anxiety Scale

HTS offers the IVR version of the LSAS. For information on the paper and pencil LSAS, please reference the following articles (HTS does not sell the paper and pencil version):

Greist, J., "The Clinical Interview," Social Phobia - Diagnosis, Assessment, and Treatment, The Guilford Press, 1995, pp. 185+.

Fresco, D.M., "The Liebowitz Social Anxiety Scale: A Comparison of the Psychometric Properties of Self-Report and Clinician-Administered Formats," Psychological Medicine, 2001, pp. 1025-1035.

The Liebowitz Social Anxiety Scale was the first clinician-administered scale to evaluate the wide range of social situations that are difficult for individuals with social phobia. The scale contains 24 items, 13 concerning performance anxiety and 11 concerning social situations. Each item is rated separately for fear (0 to 3 = none, mild, moderate, severe) and avoidance behavior (0 to 3 = never, occasionally, often, usually). Thus, the LSAS provides an overall social anxiety severity rating, and scores on 4 subscales: 1) performance fear, 2) performance avoidance, 3) social fear, and 4) social avoidance. As originally designed, the LSAS requires clinician judgment for completing the ratings and is therefore not intended for use as a self-report measure. Its wording, however, is similar to that of a self-rated instrument and it has in fact been used in this manner in pharmacological research.

The LSAS is used as an outcome measure in most pharmacological trials for social phobia, as well as in many studies of cognitive-behavioral treatment.

The clinician-administered LSAS has demonstrated good clinical utility and criterion validity. A desktop computer version of the LSAS was developed and examined in a pilot study with 12 outpatients. The correlation between clinician and computer LSAS scores at baseline was .89 and the mean score difference between the computer and clinician was not significant. The mean pre-to-post treatment change score found with the computer did not differ, either statistically or clinically from the change score found with the clinician and the amount of change indicated by the two versions of the LSAS was correlated .70.

In a larger scale follow-up study, the desktop LSAS was studied in a randomized trial of an SSRI versus placebo in social phobia. Forty-four outpatients were administered both the computer and clinician LSAS at baseline and at each of 9 follow-up visits. The internal scale consistency reliability at baseline was .94, and the mean item-to-total scale correlation was .47. The correlation between the computer and clinician at baseline was .94, and remained high throughout the study, reaching .99 by visit 6 and remaining at that level throughout the study. The mean score differences between the computer and clinician at baseline was 0.54 of a point and not significant. No significant differences were found between the amount of change detected from baseline to endpoint by the computer and the clinician. Subjects' scores on the computer version of the LSAS correlated well with clinicians' global ratings of severity, providing evidence for concurrent validity. At the end of the study, computer ratings of change from baseline showed a strong relationship with both clinician (r=.72) and patient ratings of improvement (r = .70), further supporting the concurrent validity and clinical utility of the computer LSAS.