What are the corresponding Clinical Global Impressions (CGI) scores for specific Liebowitz Social Anxiety Scale (LSAS) scores?

LSAS Score Thresholds Corresponding to Clinical Global Impressions

The optimal LSAS-SR cutoff for remission (CGI-Improvement = 1, "very much improved") is ≤35, and for the clinician-administered LSAS-CA is ≤30, with a treatment response threshold (CGI-Improvement = 2, "much improved") defined as a 28-29% reduction from baseline LSAS scores. 1

Remission Thresholds (CGI-Improvement = 1)

LSAS-SR (Self-Report Version):

• Cutoff: ≤35 points demonstrates optimal balance between sensitivity (83%) and specificity (82%) for identifying remission, defined as absence of social anxiety disorder diagnosis 1
• This threshold was validated against structured diagnostic interviews in a German sample of 311 patients who completed psychotherapeutic treatment 1

LSAS-CA (Clinician-Administered Version):

• Cutoff: ≤30 points provides the best sensitivity (88%) and specificity (87%) for remission 1
• The clinician-administered version demonstrates slightly superior diagnostic accuracy compared to the self-report version 1

Response Thresholds (CGI-Improvement = 2)

Percentage Reduction Method:

• 28% reduction for LSAS-SR from baseline yields optimal sensitivity (75%) and specificity (76%) for detecting treatment response 1
• 29% reduction for LSAS-CA from baseline provides sensitivity (83%) and specificity (80%) for response 1
• Response was defined as a 1-point reduction in structured diagnostic interview severity ratings 1

Screening and Diagnostic Thresholds

For Initial Diagnosis:

• LSAS-SR ≥30 provides optimal sensitivity and specificity for identifying any social anxiety disorder 2
• LSAS-SR ≥60 is the threshold for identifying generalized subtype social anxiety disorder specifically 2
• These cutoffs were validated in 291 patients with clinician-confirmed social anxiety disorder against 53 healthy controls 2

Clinical Application Algorithm

Step 1: Baseline Assessment

• Administer LSAS at treatment initiation (week 0) to establish baseline severity 3
• Scores ≥60 indicate generalized subtype requiring more intensive intervention 2

Step 2: Early Response Monitoring

• Reassess LSAS at week 8 of treatment 3
• Calculate percentage change from baseline: (Baseline - Current)/Baseline × 100
• ≥28-29% reduction indicates treatment response (CGI = 2) 1

Step 3: Remission Assessment

• Reassess LSAS at week 16 of treatment 3
• LSAS-SR ≤35 or LSAS-CA ≤30 indicates remission (CGI = 1) 1
• Scores between response threshold and remission threshold indicate partial response (CGI = 3)

Psychometric Considerations

Reliability:

• The LSAS demonstrates excellent internal consistency and test-retest reliability across both versions 4, 5
• The scale is normally distributed in clinical populations and sensitive to treatment effects over time 5

Subscale Limitations:

• Original fear and avoidance subscales show extremely high intercorrelations (r > 0.90) and may not provide unique information beyond the total score 6
• Factor-analytically derived subscales provide more nonredundant clinical information but are less commonly used 6

Common Clinical Pitfalls

Avoid premature response assessment:

• Full therapeutic response may require 8-12 weeks at adequate treatment doses 7
• Assessing response before 8 weeks leads to false-negative conclusions about treatment efficacy 7

Do not rely solely on absolute scores without context:

• A patient with baseline LSAS of 120 who improves to 80 (33% reduction) demonstrates response despite remaining above diagnostic threshold 1
• Conversely, a patient starting at 45 who improves to 38 (16% reduction) has not achieved response despite being near remission threshold 1

Monitor for measurement version consistency:

• LSAS-SR and LSAS-CA have different cutoffs and should not be used interchangeably for longitudinal monitoring 1
• The clinician-administered version provides slightly better diagnostic accuracy but requires more clinical time 4