PANSS: scoring, cutoffs & interpretation

01 - What it measures

What this scale measures.

The Positive and Negative Syndrome Scale (PANSS) is a 30-item, clinician-rated measure of symptom severity in schizophrenia, scored from a 30-45 minute semi-structured interview plus collateral information. It deliberately separates three things a single "psychosis severity" number blurs: positive symptoms (7 items - delusions, hallucinations, conceptual disorganisation and related), negative symptoms (7 items - blunted affect, emotional withdrawal, poor rapport and related), and general psychopathology (16 items spanning anxiety, depression, somatic concern, impulse control and insight).

Each item is rated 1 (absent) to 7 (extreme) against defined anchor points. It is the reference standard for measuring antipsychotic treatment effects in clinical trials, and in clinical practice it is used where a structured, repeatable severity measure is needed rather than a quick screen.

02 - Origin & purpose

Where it comes from.

Published by Stanley Kay, Abraham Fiszbein and Lewis Opler in Schizophrenia Bulletin in 1987, the PANSS was built from the BPRS and the Psychopathology Rating Schedule to give balanced, operationalised coverage of positive and negative syndromes - the negative syndrome having been poorly served by earlier instruments. Standardisation work in 101 patients with schizophrenia established normally distributed scale scores and formal anchor definitions for every item.

Its structured interview (SCI-PANSS), item anchors and training materials made it the common currency of antipsychotic research: regulatory trials, meta-analyses and treatment guidelines define response and remission in PANSS terms. That role is why calibration work linking PANSS scores to global clinical impressions (CGI) exists - clinicians needed to know what the numbers mean at the bedside.

03 - Scoring & cutoffs

How scoring works.

Each of the 30 items is rated 1-7. Subscale sums: positive 7-49, negative 7-49, general psychopathology 16-112; total 30-210. Because every item starts at 1, the scale floor is 30, not 0 - a detail that matters when computing percentage change (subtract 30 first, or improvements are systematically understated). There are no diagnostic cutoffs; severity benchmarks come from equipercentile linking with the CGI (Leucht et al. 2005): approximately 58 = mildly ill, 75 = moderately ill, 95 = markedly ill, 116 = severely ill. In trials, response is typically defined as a 20-25% total-score reduction (baseline-corrected) for at least minimal improvement, with 50% or greater reduction corresponding to "much improved".

04 - Validation evidence

How well it performs.

Psychometric evaluation in the standardisation samples supported internal consistency and construct validity of the three scales (Kay et al. 1987, 1988); independent replication in 100 consecutive admissions found good inter-rater reliability for the positive and negative scales but weaker agreement on general psychopathology (Peralta & Cuesta 1994).

05 - How it compares

How it compares to the alternatives.

06 - When to use it

Right tool, wrong tool.

07 - Confidence & precision

Reading the score with care.

No single agreed minimal clinically important difference exists; the CGI-linkage is the practical yardstick - a roughly 19-28% baseline-corrected reduction corresponds to "minimally improved" and 40-53% to "much improved", depending on timepoint (Leucht 2005). Two precision cautions: percentage change must be computed after subtracting the 30-point floor, and general-psychopathology ratings agree less well between raters than the positive and negative scales, so total-score movement should be read alongside subscale movement.

08 - Limitations

What it cannot tell you.

Length and training burden: a 30-45 minute interview plus formal rater training, with drift between raters well documented. The 1-7 floor complicates percentage-change arithmetic and inflates apparent baselines. The original three-factor structure is contested - five-factor models (adding disorganised and excited/hostility factors) fit better in most factor-analytic work.

General subscale inter-rater reliability is the weakest link. Scores reflect the interview window (usually the past week) and depend on patient engagement and collateral quality. Licensing costs and administration burden put it out of reach of most routine clinics.

09 - Licensing, explained

How licensing works.

The PANSS is a copyrighted instrument. Copyright is held by its authors (returned to them from Multi-Health Systems on 3 March 2025), with Mapi Research Trust acting as exclusive licensing and distribution partner - requests go through eprovide.mapi-trust.org and submitting a request is free, though licence terms and any fees depend on use. Rater training and certification are available through the PANSS Institute. Because of this, Aisel does not host the items or an online calculator. Free alternatives for dimensional psychosis ratings include the CRDPSS (APA, free) and CGI-S.

Common questions from clinicians

FAQ

References

Source literature.

1. [1]Kay SR, Fiszbein A, Opler LA The Positive and Negative Syndrome Scale (PANSS) for schizophrenia. Schizophr Bull. 1987;13(2):261-276 ↩
2. [2]Kay SR, Opler LA, Lindenmayer JP Reliability and validity of the Positive and Negative Syndrome Scale for schizophrenics. Psychiatry Res. 1988;23(1):99-110 ↩
3. [3]Peralta V, Cuesta MJ Psychometric properties of the Positive and Negative Syndrome Scale (PANSS) in schizophrenia. Psychiatry Res. 1994;53(1):31-40 ↩
4. [4]Leucht S, Kane JM, Kissling W, Hamann J, Etschel E, Engel RR What does the PANSS mean? Schizophr Res. 2005;79(2-3):231-238 ↩
5. [5]Leucht S, Kane JM, Etschel E, Kissling W, Hamann J, Engel RR Linking the PANSS, BPRS, and CGI: clinical implications. Neuropsychopharmacology. 2006;31(10):2318-2325 ↩