PANSS (Positive and Negative Syndrome Scale): 30-Item Scoring, Rescaling Rule, BPRS, and YMRS

The Positive and Negative Syndrome Scale (PANSS)—developed by Stanley R. Kay, Abraham Fiszbein, and Lewis A. Opler in 1987 (Schizophrenia Bulletin, 13[2]:261–276)—is the gold-standard 30-item clinician-administered rating instrument (ClinRO) used to measure symptom severity and antipsychotic treatment efficacy in schizophrenia and schizoaffective disorder clinical trials. Adapted from the 18-item Brief Psychiatric Rating Scale (BPRS) and the Psychopathology Rating Schedule, the PANSS provides balanced, operationalized assessment of positive symptoms, negative symptoms, and general psychopathology in psychiatry research.

How Are the 30 Items of the Kay Positive and Negative Syndrome Scale (PANSS) Scored?

The PANSS consists of 30 items, each rated by a trained clinician on a 7-point ordinal severity scale from 1 to 7 based on the 30–45 minute semi-structured Structured Clinical Interview for the PANSS (SCI-PANSS) and informant/nursing reports covering the previous 7 days:

• 1 = Absent (Notice that an asymptomatic item is scored 1, never 0)
• 2 = Minimal (questionable or suspected pathology, upper extreme of normal limits)
• 3 = Mild (clearly present symptom, but minor and does not significantly impair daily functioning)
• 4 = Moderate (distinctly serious symptom that intrudes on thinking or behavior occasionally)
• 5 = Moderate Severe (marked pathology that distinctly impairs interpersonal or daily functioning)
• 6 = Severe (gross pathology present most of the time, requiring close supervision)
• 7 = Extreme (incapacitating pathology dominating virtually all spheres of functioning)

In the classical Kay et al. (1987) scoring structure, the 30 items are divided into three subscales summing to a total PANSS score ranging from 30 to 210:

1. Positive Scale (7 items, P1–P7, Subscale Range 7–49):
   • P1 Delusions, P2 Conceptual disorganization, P3 Hallucinatory behavior, P4 Excitement, P5 Grandiosity, P6 Suspiciousness/persecution, P7 Hostility.
2. Negative Scale (7 items, N1–N7, Subscale Range 7–49):
   • N1 Blunted affect, N2 Emotional withdrawal, N3 Poor rapport, N4 Passive/apathetic social withdrawal, N5 Difficulty in abstract thinking, N6 Lack of spontaneity and flow of conversation, N7 Stereotyped thinking.
3. General Psychopathology Scale (16 items, G1–G16, Subscale Range 16–112):
   • G1 Somatic concern, G2 Anxiety, G3 Guilt feelings, G4 Tension, G5 Mannerisms and posturing, G6 Depression, G7 Motor retardation, G8 Uncooperativeness, G9 Unusual thought content, G10 Disorientation, G11 Poor attention, G12 Lack of judgment and insight, G13 Disturbance of volition, G14 Poor impulse control, G15 Preoccupation, G16 Active social avoidance.

Kay et al. also defined the Composite Index (Positive Scale score minus Negative Scale score, range -42 to +42) to classify whether positive or negative syndromes predominate in a given patient.

Why Must Percentage Change on the PANSS (30–210 Range) Be Calculated After Subtracting the 30-Point Baseline Minimum (PANSS - 30, Leucht et al.)?

Because the PANSS uses a 1–7 rating scale rather than a 0–6 scale, a completely asymptomatic individual receives a minimum total score of 30 (30 items × 1), not 0. As Stefan Leucht, Werner Kissling, and John M. Davis demonstrated in Schizophrenia Bulletin (2009, “The PANSS should be rescaled”) and equipercentile linking studies against the Clinical Global Impressions (CGI) scale (Leucht et al., 2005):

• If a biostatistician calculates percentage reduction using raw PANSS scores without subtracting 30, a patient whose score drops from 90 at baseline to 30 (complete absence of all 30 symptoms) appears to achieve only a 66.7% reduction ((90 - 30) / 90) rather than 100% symptom resolution.
• Consequently, FDA/EMA statistical analysis plans (SAPs) and peer-reviewed antipsychotic trials calculate PANSS percentage change by subtracting the 30-point floor first (replacing 1–7 with 0–6):

  Corrected % PANSS Reduction = [(PANSS_baseline - 30) - (PANSS_endpoint - 30)] / (PANSS_baseline - 30) × 100%
                              = (PANSS_baseline - PANSS_endpoint) / (PANSS_baseline - 30) × 100%

• Under Leucht et al. (2005) CGI-linking benchmarks, a 20%–25% corrected PANSS reduction corresponds to “minimally improved” on the CGI-Improvement (CGI-I = 3), whereas a 50% corrected PANSS reduction corresponds to “much improved” (CGI-I = 2). For cross-sectional severity (CGI-Severity): PANSS ≈ 58 corresponds to mildly ill, PANSS ≈ 75 to moderately ill, PANSS ≈ 95 to markedly ill, and PANSS ≈ 116 to severely ill.

How Do the Marder 5-Factor PANSS Model, 18-Item BPRS, and 11-Item Young Mania Rating Scale (YMRS) Support Psychiatric Clinical Trials?

In contemporary psychiatric clinical trials and electronic clinical outcome assessment (eCOA) platforms, the classical 3-subscale PANSS is routinely analyzed alongside the Marder 5-Factor Model, the 18-item BPRS, and—for bipolar I mania trials—the 11-item Young Mania Rating Scale (YMRS):

• Marder 5-Factor PANSS Model (Marder, Davis, & Chouinard, 1997): Factor analyses across thousands of trial participants showed that the 30 PANSS items segregate more cleanly into five orthogonal dimensions than three subscales:
  1. Positive Factor (8 items): P1, P3, P5, P6, N7, G1, G9, G12
  2. Negative Factor (7 items): N1, N2, N3, N4, N6, G7, G16
  3. Disorganized Thought Factor (7 items): P2, N5, G5, G10, G11, G13, G15
  4. Uncontrolled Hostility / Excitement Factor (4 items): P4, P7, G8, G14
  5. Anxiety / Depression Factor (4 items): G2, G3, G4, G6
• Brief Psychiatric Rating Scale (BPRS; Overall & Gorham, 1962): The 18-item clinician-rated precursor to the PANSS (1–7 rating scale, total score 18–126, requiring an 18-point floor subtraction for percentage change; or scored 0–6 for a 0–108 range). All 18 BPRS items are embedded within the 30-item PANSS, allowing trials to derive a “PANSS-derived BPRS” score automatically.
• Young Mania Rating Scale (YMRS; Young, Biggs, Ziegler, & Meyer, 1978): The gold-standard 11-item clinician-rated primary endpoint (total range 0–60) for acute bipolar I mania trials (complementing self-report hypomania screens such as the HCL-32 and depression scales such as the Hamilton Depression Rating Scale). Seven YMRS items (elevated mood, increased motor activity/energy, sexual interest, sleep, language-thought disorder, appearance, and insight) are scored 0–4, while four core items are double-weighted and scored 0–8 (Item 5: Irritability, Item 6: Speech rate and amount, Item 8: Content of thought, and Item 9: Disruptive-aggressive behavior) so clinicians can still assign accurate severity when manic agitation prevents full verbal cooperation. Standard YMRS remission in bipolar mania trials is defined as YMRS ≤ 12 (or ≤8 for strict euthymia).

Instrument	Primary Indication	Items & Item Scaling	Total Score Range	Rescaling / Remission Benchmark
PANSS (Kay et al., 1987)	Schizophrenia & schizoaffective disorder	30 items (7 P, 7 N, 16 G) rated 1–7 via SCI-PANSS	30–210	Subtract 30 before % change; Andreasen remission criteria require 8 core items (P1, P2, P3, N1, N4, N6, G5, G9) all scored ≤3 for ≥6 months
BPRS-18 (Overall & Gorham, 1962)	Broad psychotic & severe psychiatric disorders	18 items rated 1–7 (or 0–6)	18–126 (or 0–108)	Subtract 18 when using 1–7 scaling; fully derivable as a subset of the 30 PANSS items
YMRS (Young et al., 1978)	Acute bipolar I mania	11 items: 7 items rated 0–4; 4 core items (5, 6, 8, 9) rated 0–8	0–60	Response: ≥50% reduction from baseline; Remission: YMRS ≤ 12

Frequently Asked Questions

How are the 30 items of the Kay Positive and Negative Syndrome Scale (PANSS: 7 Positive, 7 Negative, 16 General Psychopathology, rated 1–7) scored?

Using the Structured Clinical Interview for the PANSS (SCI-PANSS), a clinician rates 7 Positive items (P1–P7, range 7–49), 7 Negative items (N1–N7, range 7–49), and 16 General Psychopathology items (G1–G16, range 16–112) from 1 (Absent) to 7 (Extreme), producing a total score between 30 and 210.

Why must percentage change on the PANSS (30–210 range) be calculated after subtracting the 30-point baseline minimum (PANSS - 30, Leucht et al.)?

Because the lowest possible rating on each of the 30 PANSS items is 1 (Absent), a patient with zero symptoms scores 30, not 0. Failing to subtract 30 from the denominator (PANSS_baseline - 30) mathematically underestimates percentage symptom improvement by up to one-third to one-half.

How do the Marder 5-Factor PANSS model, 18-item BPRS, and 11-item Young Mania Rating Scale (YMRS) support psychiatric clinical trials?

The Marder 5-factor model reorganizes the 30 PANSS items into Positive, Negative, Disorganized Thought, Hostility/Excitement, and Anxiety/Depression factors to isolate domain-specific drug effects; the 18-item BPRS provides a rapid psychosis severity score that can be extracted directly from the PANSS; and the 11-item YMRS (0–60, with double-weighted 0–8 items for irritability, speech, thought content, and aggression) serves as the primary clinician-rated endpoint in acute mania trials.