Bush-Francis Catatonia Rating Scale (BFCRS)

Gregory Bush, Max Fink, Georgios Petrides, Francis Dowling, Andrew Francis · 1996

All twenty-three catatonia signs with their published anchors — the field's standard instrument, and one it already teaches from video.

The BFCRS is the instrument catatonia research runs on, and almost every sign in it is something you have to see: posturing, staring, grimacing, stereotypy, mannerisms, waxy flexibility. Its authors' own training resource pairs every item with an example video, so a video annotation workflow is not an adaptation of this scale — it is how the field already teaches it.

Twenty-three items, two rating patterns. Seventeen are graded 0 to 3 with item-specific anchors. Six are binary, scored 0 or 3 only: waxy flexibility, mitgehen, gegenhalten, ambitendency, grasp reflex and perseveration. Their presence is what matters; there is no meaningful "mild gegenhalten". Note that combativeness and autonomic abnormality are graded 0 to 3 despite often being listed among the binary items in secondary sources.

The screening instrument is the first fourteen items (BFCSI): excitement, immobility/stupor, mutism, staring, posturing/catalepsy, grimacing, echopraxia/echolalia, stereotypy, mannerisms, verbigeration, rigidity, negativism, waxy flexibility and withdrawal. Two or more present is the commonly used screening threshold — cite that to the reliability literature, not to the manual, and note it is a screen rather than a diagnosis.

This scale requires the examination, not just an interview. Mitgehen, gegenhalten, grasp reflex, rigidity and waxy flexibility are elicited by the examiner's hands; autonomic abnormality needs vital signs. They are rateable from a recording of the standardized examination and not from a recording of a conversation. If your footage does not contain the manoeuvre, do not infer the sign — the manual's own rule is that an item you are unsure of is rated 0.

Declare which window you rated. The published form distinguishes a state examination from an interval examination over a stated number of hours, with a floor of about five minutes or the duration of the assessment. On a timeline this becomes the span you annotate, so tag the assessment window and say which kind it is; a state and an interval rating of the same patient are not the same measurement.

Some items are audible, not visual. Mutism, verbigeration, echolalia and stereotyped speech show up in a transcript with timings, which is what makes them the few items an AI agent can genuinely help with here. The motor signs are for your eyes only, and the curated skills say so.

Domains (23)

ExcitementBEX

Extreme hyperactivity, constant motor unrest that appears non-purposeful; not attributable to akathisia or goal-directed agitation.

Curated skill
Immobility or StuporBIM

Extreme hypoactivity, immobility, minimal responsiveness to stimuli.

Curated skill
MutismBMU

Verbal unresponsiveness — from incomprehensible whisper through markedly reduced output to no speech at all.

Curated skill
StaringBST

Fixed gaze, little or no visual scanning of the environment, reduced blinking.

Curated skill
Posturing or CatalepsyBPO

Spontaneous maintenance of a posture, including mundane ones, held well beyond what would be comfortable.

Curated skill
GrimacingBGR

Maintenance of odd facial expressions.

Curated skill
Echopraxia or EcholaliaBEC

Mimicking the examiner's movements or repeating their speech.

Curated skill
StereotypyBSE

Repetitive, non-goal-directed motor activity such as finger play or repeatedly touching or rubbing.

Curated skill
MannerismsBMA

Odd, purposeful movements — hopping or walking on tiptoe, saluting passers-by, exaggerated caricature of a normal action.

Curated skill
VerbigerationBVE

Repetition of phrases or sentences, like a scratched record.

Curated skill
RigidityBRI

Maintenance of a rigid position despite efforts to move the person; exclude cogwheeling or tremor if present.

Curated skill
NegativismBNE

Apparently motiveless resistance to instructions or attempts to move or examine; contrary behaviour to the request.

Curated skill
Waxy FlexibilityBWX

During repositioning the person offers initial resistance before allowing the limb to be moved, like bending a candle. Present or absent only.

Curated skill
WithdrawalBWI

Refusal to eat, drink or make eye contact.

Curated skill
ImpulsivityBIP

Sudden inappropriate behaviour without provocation, afterwards unexplained by the person.

Curated skill
Automatic ObedienceBAO

Exaggerated cooperation with the examiner's request, or repeated movement when requested once.

Curated skill
MitgehenBMG

"Anglepoise lamp" raising of the arm in response to light finger pressure, despite instruction to resist. Present or absent only.

Curated skill
GegenhaltenBGH

Resistance to passive movement that is proportional to the force applied, seemingly automatic rather than willed. Present or absent only.

Curated skill
AmbitendencyBAM

The person appears motorically stuck in indecisive, hesitant movement. Present or absent only.

Curated skill
Grasp ReflexBGP

Per the standardized neurological examination — the hand closes on stimulation of the palm. Present or absent only.

Curated skill
PerseverationBPE

Repeatedly returning to the same topic or persisting with a movement after it is no longer called for. Present or absent only.

Curated skill
CombativenessBCO

Belligerence or aggression, usually in an undirected way and without explanation.

Curated skill
Autonomic AbnormalityBAU

Abnormality of temperature, blood pressure, pulse, respiratory rate, or inappropriate sweating. Requires observations, not appearance.

Curated skill
Bush G, Fink M, Petrides G, Dowling F, Francis A. Catatonia. I. Rating scale and standardized examination. Acta Psychiatr Scand. 1996;93(2):129-136. doi:10.1111/j.1600-0447.1996.tb09814.x

Freely reproducible with citation: the complete scale, its anchors and the training manual are openly distributed by the authors through the University of Rochester Medical Center, which also publishes per-item example videos. The journal article itself remains publisher-copyrighted — cite Bush et al. (1996) and the URMC training resource. Per-level descriptions here are condensed for a one-line rating control; rate against the manual.