Screening and measurement scales for social anxiety

Which scale for which job

There are two jobs. Screening asks: could this person have social anxiety disorder? Measuring asks: how severe is it, and is treatment working? Short scales screen. Longer scales measure. No scale makes a diagnosis. That needs a clinical interview.

NICE CG159 (England, 2013) puts it like this: if you suspect social anxiety disorder, use the 3-item Mini-SPIN, or ask whether the person avoids social situations and whether they feel fearful or embarrassed in them. A Mini-SPIN score of 6 or more, or a “yes” to either question, should lead to a full assessment. For that assessment and for tracking change, NICE names the SPIN or the LSAS.

Notes on the table:

• The LSAS was first published by Michael Liebowitz in 1987. The self-report and clinician versions behave in much the same way, and the same cut-offs of 30 and 60 held for the self-report version in 291 patients and 53 controls.
• The SIAS and SPS are companion scales. In 577 patients, they measured related but separate things, so use both if you want the full picture.
• The SIAS-6 and SPS-6 were built from 456 patients in five Australian treatment trials. They tracked change as well as the full scales did.
• Cut-offs for the SIAS, SPS and SPAI vary between studies, and none was checked for this page, so none is printed here.

What a positive screen means, and what it doesn’t

Screening scores are often read as diagnoses. They are not.

In the Mini-SPIN study, carried out in a US managed-care population, a score of 6 or more caught 88.7% of people who had the disorder (sensitivity) and correctly cleared 90.0% of people who did not (specificity). But the positive predictive value was only 52.5%. About half of the people who screened positive did not have social anxiety disorder. The negative predictive value was 98.5%: a negative screen is very reassuring.

In plain terms for the person in front of you: “A high score means it is worth a proper assessment. It is close to a coin toss whether that assessment will confirm the disorder.”

This matters when screening scores turn into headlines:

• A survey of 6,825 people aged 16 to 29 in seven countries found 36% above a SIAS screening threshold. That is a self-report cut-off, not a diagnosis rate.
• In about 450,000 Finnish teenagers, the share above the Mini-SPIN cut-off roughly doubled between 2013/15 and 2021, reaching 47% of girls. Again, a symptom screen, not diagnoses.

Positive predictive value falls when a disorder is less common in the group you screen. A cut-off tuned in a clinic will flag many more false positives in a general population or a school.

Who owns these scales, and can they go on a website?

None of the scales below is public domain or under an open licence such as Creative Commons. Most are free for clinicians to use with their own patients or for research. That is not the same as permission to publish them or to run them as an online self-test. This site therefore does not reproduce any items or offer interactive scoring.

Source links for each row are listed under Sources at the bottom of this page. Where the terms come from an author’s email quoted by a third party, the original author page could not be found.

Where to get them legitimately

• DSM-5 Severity Measure: free PDF from the American Psychiatric Association’s website, with scoring instructions. For anything beyond clinical or research use, use the APA permissions form.
• SPIN and Mini-SPIN: ask the copyright holder, Jonathan Davidson, for permission. The copyright notice gives a contact address.
• LSAS: the scale appears in the original 1987 publication. Contact Michael Liebowitz for commercial or other non-clinical use.
• SIAS and SPS: printed in the 1998 article by Mattick and Clarke. Acknowledge the source.
• SPAI: from the test publisher, if still available in your country.

If you work in a service that wants a digital version (for example, in a patient portal), get written permission first. The terms above cover use with your own patients, not building it into a product.

Using scores well in therapy

• Give the same measure at the start, during and at the end of treatment. Session-by-session measures help spot sudden gains and stalls.
• Use a self-report measure and your own rating. Self-report and clinician ratings can disagree, especially when someone wants to please you or plays down symptoms.
• Ask about the situations behind a high score. Two people with the same LSAS total can fear completely different things, and the exposure plan depends on that.
• Remember what the scales miss. None asks directly about fear of offending others (taijin kyofusho), and in autistic people, high scores may partly reflect social communication differences. See adapting treatment.