Rating scales
Score the Brief Psychiatric Rating Scale: eighteen items on whichever numbering your form uses — 0–6 or the original 1–7 — with the five factor subscales, the CGI-anchored severity band, and an approximate PANSS equivalent.
Free · No sign-up · Runs entirely in your browser
Rate each item from the interview and the period it covers. Anchors run from not present to extremely severe.
Switching this relabels the buttons and the total; it never changes a rating. The two numberings describe the same seven anchors, so the same patient scores 18 points higher on the original 1–7 form — which is why a BPRS total quoted without its numbering is ambiguous.
Preoccupation with physical health, whether or not medically grounded
Reported worry, apprehension or fear about the present or future
Deficient relating to the interviewer and the immediate situation
Disconnected, tangential or incoherent speech — rated on form, not content
Self-reproach or remorse over past acts, real or imagined
Observed physical and motor manifestations of nervousness
Odd, stylised or unnatural movement or posture
Exaggerated self-opinion; inflated conviction of ability or identity
Despondency, sadness or discouragement
Animosity, contempt or belligerence toward others
Mistrust, from wariness through to frank persecutory delusion
Perception without an external stimulus
Slowed or reduced movement and speech
Resistance, guardedness or refusal within the interview
Unusual, odd or delusional beliefs — rated on content, not form
Reduced emotional tone; flat voice, impassive expression
Heightened emotional tone, agitation, increased reactivity
Confusion about person, place or time
Total (0–6 numbering)
0 / 108
Floor is 0 on this numbering
Same score, 1–7 numbering
18
Identical ratings, relabelled
Severity band
Not or borderline ill
Leucht et al., 2005 (CGI-anchored) · reference band, not a diagnosis
Approx. PANSS total
31
Leucht et al., 2013 · population estimate, not a PANSS rating
Factor subscales — Shafer (2005) meta-analysis of 26 factor analyses
Affect
0 / 24
4 items
Positive symptoms
0 / 24
4 items
Negative symptoms
0 / 18
3 items
Resistance
0 / 18
3 items
Activation
0 / 18
3 items
These five cover seventeen of the eighteen items. Item 18, disorientation (currently 0), loads on no factor in the meta-analysis, so the subtotals are not expected to add up to the total. The first four factors are the stable core; activation emerged clearly only in samples of patients with schizophrenia.
Plain text: every item with its code, both totals with their numbering named, all five factor subtotals, and the PANSS estimate or the reason there isn’t one.
Items follow Overall & Gorham (1962); factors follow Shafer (2005); severity bands follow Leucht et al. (2005); the PANSS correspondence follows Leucht et al. (2013). This is a scoring aid for education and research support, not a diagnosis. The BPRS is a clinician-rated instrument and its reliability depends on rater training. Rate the items above, or load the example, to see the score update.
Reference · for the curious
The Brief Psychiatric Rating Scale is the oldest broad measure of psychopathology still in wide use. Overall and Gorham published it in 1962, and its eighteen items span psychotic and affective illness in a way most later instruments do not. It is also in the public domain, which is the practical reason it persists alongside the licensed PANSS.
Start here, because it is the most common way a BPRS score gets misread. Overall and Gorham numbered the seven anchor points 1 to 7, which puts the floor of an eighteen-item scale at 18 and its ceiling at 126. Most modern use renumbers the same seven anchors 0 to 6, giving 0 to 108. The ratings are identical. Only the labels differ — by exactly eighteen points, one for each item.
So a reported “BPRS total of 31” is ambiguous. On the original numbering it is a mildly ill patient; on the modern numbering it is a moderately ill one. That is a two-band difference from a labelling convention nobody states. The calculator above shows both totals at once for this reason, and any BPRS score you publish should name its numbering explicitly.
The BPRS has no official severity band table of the kind the PHQ-9 or MADRS carry. The usual reference points come from Leucht and colleagues (2005), who linked BPRS totals to Clinical Global Impression ratings: “mildly ill” corresponded to a total of about 31, “moderately ill” to about 41, and “markedly ill” to about 53. Those are on the original 1–7 numbering — roughly 13, 23 and 35 on the modern one. If you report a BPRS band, cite where the band came from.
Shafer's (2005) meta-analysis pooled 26 factor analyses of the BPRS across 17,620 patients and supported five subscales: Affect (anxiety, guilt, depressive mood, somatic concern); Positive Symptoms (unusual thought content, conceptual disorganization, hallucinatory behaviour, grandiosity); Negative Symptoms (blunted affect, emotional withdrawal, motor retardation); Resistance (hostility, uncooperativeness, suspiciousness); and Activation (excitement, tension, mannerisms and posturing).
The first four are the stable core. Activation emerged clearly only in samples of patients with schizophrenia, so treat it as the least portable of the five. And note the arithmetic: the five factors between them claim seventeen of the eighteen items. Disorientation belongs to none of them, which is why the subtotals in the tool do not add up to the total. That is a property of the scale, not a bug in the sum.
Two items are also rated on a different basis from the rest, and confusing them is a classic rater error. Conceptual disorganization is rated on the form of speech — whether it hangs together. Unusual thought content is rated on its content — whether the belief expressed is odd. A person can be floridly delusional in perfectly organised sentences, or incoherent without expressing any unusual belief. Our comparison of BPRS versus PANSS works through the item-level overlap, and positive formal thought disorder covers the form-versus-content distinction in more depth.
Leucht and colleagues (2013) applied equipercentile linking to 3,767 patients with simultaneous BPRS and PANSS ratings, and found that BPRS totals of 18, 30, 40 and 50 correspond to PANSS totals of roughly 31, 55, 73 and 90. The calculator interpolates between those published anchors, and deliberately returns nothing outside them: extending the line past 50 would produce a number the paper does not support.
Read the conversion for what it is. Equipercentile linking matches distributions — it tells you which PANSS total sits at the same percentile as a given BPRS total in a trial population. It does not make the two instruments interchangeable for an individual patient. The BPRS carries three negative-symptom items against the PANSS's seven, so two people with the same BPRS total can have substantially different PANSS profiles, and a converted score should never be presented as though the other scale had actually been administered.
This matters commercially as well as methodologically: the PANSS is licensed and the BPRS is not, so the BPRS is the instrument you can host, translate and redistribute freely. If the PANSS is what your protocol calls for but licensing is the obstacle, the BPRS covers most of the same ground for positive, affective and resistance symptoms. Both are described in the Tagaroo Scale Library.
Eleven of the eighteen items can be grounded in what the subject says. The rest need observation — mannerisms and posturing, motor retardation, blunted affect and tension are rated on how the person appears and sounds, not on the words alone — and audio or video is the honest minimum for those. Disorientation usually requires having asked.
Decide in advance how an unelicited item is handled, because the BPRS is a semi-structured rating without a fixed interview: an item nobody asked about is not the same as an item rated zero. Our guide to the BPRS's items and factor structure covers the anchors, and writing annotation guidelines that hold up covers making those decisions explicit before anyone codes. With two raters, compute agreement per item as well as on the total using the inter-rater reliability calculator — item-level agreement on this scale is markedly weaker than total-score agreement, and only the item view shows you which anchors need rewriting.
Total-score inter-rater reliability is generally good with trained raters, and the scale's longevity rests on that. Item-level reliability is considerably more variable, particularly for the items that require inference about internal states rather than description of behaviour. The scale predates modern psychometric practice — it was assembled to be brief and clinically useful, not to be factorially clean — so its dimensionality has been re-litigated repeatedly, which is exactly what Shafer's meta-analysis set out to settle. Report your rater training and your agreement figures; on this instrument they are not boilerplate.
The BPRS is in the public domain. Overall and Gorham published it in Psychological Reports in 1962; it requires no permission or fee to reproduce, translate or distribute, and should be cited to them. This is a substantive advantage over the PANSS for any project that needs to publish its instrument, translate it, or share a coding scheme built on it.
This is a scoring aid for education and research support. It does not diagnose, it cannot distinguish a primary psychotic disorder from psychotic features of a mood episode or an organic cause, and it does not replace clinical assessment. Everything is computed in your browser; no ratings are transmitted or stored.
Both, and that is the scale's standing trap. Overall and Gorham numbered the seven points 1–7 in 1962, which puts the floor of an 18-item scale at 18 and the ceiling at 126. Most modern use renumbers the same seven points 0–6, giving a range of 0–108. The ratings are identical; only the labels differ. A bare "BPRS total of 31" is therefore ambiguous by 18 points — it is either a mildly ill patient on the original numbering or a moderately ill one on the modern numbering. This calculator shows both totals at once, and any BPRS score you report or read should say which numbering it uses.
Leucht and colleagues (2005) linked BPRS totals to Clinical Global Impression ratings and found that "mildly ill" corresponded to a BPRS total of about 31, "moderately ill" to about 41, and "markedly ill" to about 53. Those figures are on the original 1–7 numbering, so on the 0–6 numbering they are roughly 13, 23 and 35. The BPRS has no single official severity band table of the kind the PHQ-9 and MADRS have, which is why these CGI-anchored values are the usual reference points — and why a reported band should always name its source.
Shafer's (2005) meta-analysis of 26 factor analyses (N = 17,620) supported five subscales: Affect (anxiety, guilt, depressive mood, somatic concern); Positive Symptoms (unusual thought content, conceptual disorganization, hallucinatory behaviour, grandiosity); Negative Symptoms (blunted affect, emotional withdrawal, motor retardation); Resistance (hostility, uncooperativeness, suspiciousness); and Activation (excitement, tension, mannerisms and posturing). The first four form the core structure; activation emerged clearly only in samples of patients with schizophrenia. Together the five account for seventeen of the eighteen items — disorientation loads on none of them, so the subscales do not add up to the total.
Leucht and colleagues (2013) used equipercentile linking on 3,767 patients with simultaneous ratings and found that BPRS totals of 18, 30, 40 and 50 correspond to PANSS totals of roughly 31, 55, 73 and 90. The linking is a rank-order correspondence between distributions, not a formula: it tells you which PANSS total sits at the same percentile as a given BPRS total in a trial population, and it does not make the two scales interchangeable for an individual patient. The BPRS contains no negative-symptom items beyond three, while the PANSS devotes seven to them, so two patients with the same BPRS total can have quite different PANSS profiles.
Conceptual disorganization is rated on the form of speech; unusual thought content is rated on its content. Confusing the two is a classic BPRS rater error. Conceptual disorganization asks whether the speech hangs together — is it tangential, disconnected, incoherent? Unusual thought content asks whether the belief being expressed is odd or delusional, regardless of how fluently it is said. The two come apart in both directions: a person can voice a florid persecutory delusion in perfectly organised, grammatical sentences, which scores on unusual thought content and not on conceptual disorganization; and a person can be markedly incoherent without expressing any unusual belief at all, which scores the other way round. Rate them independently, and note that only conceptual disorganization loads on the positive-symptom factor alongside unusual thought content in Shafer's (2005) structure.
The BPRS is in the public domain — Overall and Gorham published it in Psychological Reports in 1962 — and requires no permission or fee, which is a practical argument for it. The PANSS is a licensed instrument distributed by a commercial publisher. The trade-off is coverage: the PANSS is longer (30 items) and much richer on negative symptoms, so it is the conventional choice in schizophrenia trials, while the BPRS is shorter, faster and spans psychotic and affective illness. If cost or redistribution is the constraint, the BPRS is the free instrument that covers most of the same ground.
Written by Enrique Gutiérrez, PhD (Computer Science) — founder of Tagaroo and Associate Professor of Computer Science, working on inter-rater reliability, measurement and annotation methodology (ORCID).
How this page is checked. Scoring rules, item ranges and thresholds are transcribed from the instrument's cited primary sources and covered by automated tests that reproduce each paper's own worked examples. This page has not been reviewed by a licensed clinician, and it is a scoring aid for education and research support rather than a clinical decision tool.
Last verified: 29 July 2026. Formulas, thresholds and cited figures on this page were checked against their original sources on that date. Every calculation runs in your browser; nothing you enter is transmitted or stored.
Tagaroo lets you score the BPRS against a transcript, with each item rating linked to the passage that supports it — including the two items rated on the form of speech rather than its content.
BPRS calculator · tagaroo.ai/materials/bprs-calculator · figures verified against primary sources 2026-07-29. Educational scoring aid, not a diagnosis, and not reviewed by a licensed clinician.