Notice. Illustration and research prototype. Not clinically validated, not a medical device, not for patient care. Example data are fictional.
Live demonstration. The sliders and test cases work — every score recomputes in your browser. The data is generated, no patient is involved, and nothing here has been verified by a clinician.

workload.yaml

Configuration · workload.yaml · 283 lines · editable in place

NURSING WORKLOAD AND ALLOCATION ============================================================================ Idea 73. Staffing to predicted acuity rather than to bed count.

THE PROBLEM WITH BED COUNT -------------------------- "Six beds, three nurses" says nothing about whether the shift is survivable. Two ventilated patients on renal replacement and four awaiting discharge is a different shift from six patients mid-resuscitation, and the bed count is identical.

So workload is scored from what is actually being done to each patient — the interventions, their frequency, and the hours they consume.

WHAT MAKES THIS ENGINE DIFFERENT FROM THE CLINICAL ONES ------------------------------------------------------- Every other engine here advises on a patient. This one describes a shift, and it is used by people who are already overloaded. That changes two things:

IT MUST NOT MORALISE. A unit that is under-staffed knows it. Telling the sister that her ratio is unsafe is not information, it is an accusation she can do nothing with. The engine states the arithmetic and stops.

IT MUST NOT BECOME A PRODUCTIVITY INSTRUMENT. Workload scores computed per nurse become performance metrics, and then nurses under-record. The engine scores PATIENTS, never nurses, and there is no per-nurse output.

THE HONEST LIMIT ---------------- This describes demand. It cannot create staff. Where demand exceeds the roster, the engine says so plainly and lists what would have to be deferred — because an unstaffable shift is a decision about what will not be done, and making that decision explicit is better than discovering it at 04:00.

WHAT THESE WEIGHTS ARE, AND WHAT THEY ARE NOT --------------------------------------------- They are a flat minutes-per-activity table, invented for this build as a starting point. They are NOT derived from NAS or TISS-28, and an earlier version of this file wrongly said they were.

The published instruments work differently and are not interchangeable with this: NAS scores 23 items as percentages of one nurse's time, summing to a possible 177%. TISS-28 scores 28 items in TISS points, one point being roughly 10.6 minutes of nursing time. Either is a validated instrument; this table is not.

A unit that already scores NAS or TISS-28 has a validated instrument, and this table adds nothing to it. This exists for a unit that scores neither, and every figure must be re-timed locally before it carries any weight in a staffing argument.

Ratio references below are separate and are cited where used.

meta
version0.1.0
ownerUNASSIGNED — set to the responsible nurse lead
verified bynot set
boundary
Describes demand. Cannot create staff. Scores patients, never nurses, and
produces no per-nurse output of any kind.
ratio framing
The published ratios are aspirational targets, not minimum standards, and
not a judgement on a unit that does not meet them. ISCCM's own guidance
discusses them alongside explicit adaptation for resource-limited
settings.
ratio citation status
UNVERIFIED — the specific ISCCM document, year and page are not yet
confirmed. Fill these in before the ratios are used in any staffing
argument.
weights note
Minutes per shift are illustrative and were invented for this build. They
are NOT derived from NAS or TISS-28. A unit already scoring either has a
validated instrument, and this table adds nothing to it. They must be
re-timed against this unit's own practice before any use; nursing time is
the most local quantity in the whole build.
shift minutes720
12-hour shift
activities
- idbaseline_care
---- baseline --------------------------------------------------------
labelBaseline care, observation, and documentation
minutes120
appliesevery occupied bed
noteThe floor. No patient costs less than this.
- idinvasive_ventilation
---- respiratory -----------------------------------------------------
labelInvasive ventilation
minutes90
appliesventilated
- idniv
labelNon-invasive ventilation
minutes105
applieson NIV
note
Costed here ABOVE invasive ventilation, on the reasoning that a poorly
tolerated mask needs near-continuous adjustment and reassurance while a
sedated intubated patient does not.
disagreement
This inverts both published instrumentsNAS and TISS-28 weight invasive
ventilatory support at or above non-invasive. The inversion is a
deliberate local judgement, not a finding, and it is flagged rather than
buried because it is the single weight in this table most likely to be
wrong. A unit timing its own NIV care would overwrite this first.
- idsuctioning_frequent
labelFrequent suctioning
minutes45
appliessecretions requiring more than 2-hourly suction
- idprone_positioning
labelProne positioning
minutes120
appliesproned
noteNeeds 4-5 staff for each turn. Consumes the whole unit, twice.
- idvasopressor_titration
---- circulation -----------------------------------------------------
labelVasopressor titration
minutes60
applieson vasopressors
- idmultiple_infusions
labelFour or more concurrent infusions
minutes40
applies4+ infusions
- idhaemodynamic_monitoring
labelInvasive haemodynamic monitoring
minutes30
appliesarterial line or cardiac output monitoring
- idcrrt
---- renal -----------------------------------------------------------
labelContinuous renal replacement therapy
minutes100
applieson CRRT
- idhourly_urine
labelHourly urine measurement
minutes20
applieshourly output charting
- idagitation_management
---- neurological ----------------------------------------------------
labelAgitated or delirious patient
minutes130
appliesRASS +2 or above, or CAM-ICU positive
note
Among the heaviest items in the whole list, and the one most often
omitted from workload tools because it produces no procedure record.
- idneuro_observations
labelHourly neurological observation
minutes40
applieshourly GCS and pupils
- idfull_dependency
---- mobility and skin -----------------------------------------------
labelFully dependent for turning and hygiene
minutes90
appliesfully dependent
- idtwo_hourly_turns
labelTwo-hourly repositioning
minutes60
appliespressure injury risk
- idmobilisation
labelEarly mobilisation session
minutes45
appliesmobilising
note
Counted honestly. It is the activity most often skipped under pressure,
and pretending it is free is why.
- idtransport_offunit
---- procedures and transfers ----------------------------------------
labelOff-unit transfer (imaging, theatre)
minutes90
appliesleaving the unit
noteRemoves a nurse from the unit entirely for the duration.
- idprocedure_assist
labelAssisting a bedside procedure
minutes45
appliesline, drain, or bronchoscopy today
- idadmission
labelNew admission
minutes120
appliesadmitted this shift
- iddischarge_transfer
labelDischarge or ward transfer
minutes60
appliesleaving this shift
- idfamily_intensive
---- family and end of life ------------------------------------------
labelFamily requiring extended communication
minutes60
appliesfamily in crisis, or breaking bad news
- idend_of_life
labelEnd-of-life care
minutes150
applieswithdrawal or actively dying
note
The heaviest single item, and the one no bed-count model captures at
all. A dying patient is not a low-acuity patient.
- idcontact_isolation
---- isolation -------------------------------------------------------
labelContact isolation
minutes45
appliesisolation precautions
noteEvery entry and exit costs time. It compounds with everything else.
ratios
- idlevel3
These are ASPIRATIONAL TARGETS, and stating them without that framing would misrepresent the source. ISCCM's planning and staffing recommendations discuss these alongside explicit acknowledgement that Indian units frequently cannot meet them, and include adaptation for resource-limited settings. Presented flat, they become ratio dogma with an ISCCM label on it, which is the opposite of what the source argues. The citation below is deliberately marked unverified: the specific document, year and page must be filled in by the unit before these are used in any staffing argument. isccm_reference.yaml shows the standard this file has not yet met.
labelLevel 3 — ventilated or multi-organ support
nurse to patient1:1
sourceISCCM ICU planning and staffing recommendations — citation unverified
- idlevel2
labelLevel 2 — single organ support
nurse to patient1:2
sourceISCCM ICU planning and staffing recommendations — citation unverified
- idlevel1
labelLevel 1 — enhanced observation
nurse to patient1:3
note
Several Indian documents state 14 for Level 1 / HDU. Confirm which the
unit works to before relying on this.
sourceISCCM ICU planning and staffing recommendations — citation unverified
deferred first
- idmobilisation
whatEarly mobilisation
consequenceDelirium and ICU-acquired weakness — the harm arrives in days, not hours
- idtwo_hourly_turns
whatRepositioning frequency
consequencePressure injury — measurable, attributable, and preventable
- idfamily_intensive
whatFamily communication
consequenceComplaints, and decisions made without the family present
- idoral_care
whatOral care frequency
consequenceVentilator-associated pneumonia
- iddocumentation
whatContemporaneous documentation
consequenceWritten at the end of the shift from memory, or not at all