Deciding how many nurses a unit needs on a given shift is done in one of three ways: a fixed ratio, a budgeted average of nursing hours, or a measure of how sick the patients actually are. The third approach is acuity-based staffing, and the tools that produce it are patient classification or acuity systems.
The appeal is obvious. Twelve stable patients and twelve deteriorating ones are not the same work, and any method that ignores the difference will be wrong twice: overstaffed on the quiet shift, dangerous on the bad one.
The three measures, and what each hides
A ratio is the number of patients assigned to one nurse. It is the measure legislation uses, because it is the only one a nurse can verify at the bedside in the moment. What it hides is variation in how much care those patients need.
Nursing hours per patient day is a budgeting measure: productive hours worked by nursing staff with direct care responsibility, divided by patient days, calculated per unit per month. The national quality indicator set distinguishes registered nurse hours from total nursing hours including practical nurses and assistants. What it hides is distribution. A month can hit its target while individual shifts run severely short, which is precisely the gap the strongest staffing study exploited by measuring exposure to shifts eight or more hours below the unit's own plan. See nurse staffing and patient outcomes.
Acuity adjusts for need. What it hides is the judgement inside the tool.
How an acuity system works
A classification system scores each patient on dimensions such as dependency for activities of daily living, monitoring frequency, medication and infusion complexity, wound or device care, cognitive status and fall risk, and psychosocial or family demands. Scores roll up to a unit workload figure, which converts into a required number of care hours, which converts into a nurse count for the shift.
Modern systems draw much of this from the electronic record rather than from a paper form, which makes them cheaper to run and easier to trust less: a score assembled from documentation reflects what was charted, and charting is the first thing that suffers when a unit is short.
The professional association's staffing principles favour this direction over fixed ratios, on the grounds that appropriate staffing has to account for patient acuity and intensity alongside nurse experience, education and skill mix, and that electronic classification systems make those factors visible.
Where it goes wrong
Acuity-based staffing is the most defensible approach in principle and the most manipulable in practice, for a plain reason: the organisation that pays for the staffing usually chooses and configures the tool that determines how much staffing is required.
Four failure modes recur.
The scoring can be recalibrated. Adjusting the hours implied by a given score changes required staffing without changing any patient, and without any visible decision.
The output can be advisory. A system that recommends six nurses and a manager who staffs four produce a documented recommendation and a short shift.
The measure can lag. Acuity assessed at midnight does not describe a unit at four in the morning after two admissions and a deterioration.
The invisible work is missing. Precepting a new graduate, orienting an agency nurse, covering a break, and sitting with a distressed family are real hours that most tools do not count.
Why it is now a bargaining subject
Acuity systems moved from an operational detail to a contract term because the newer ones are algorithmic, and an algorithm that determines staffing without any identifiable person accountable for the judgement is a different proposition from a manager with a form.
The 2026 nursing contracts in New York were the first to contain limits on how artificial intelligence may be used in nursing work. The dispute was not about software existing on the unit. It was about who decides when the system says four nurses and the nurses on the floor say six. See nurse strikes and bargaining.
Expect the question to appear in interviews as well as in negotiations. A candidate asking which acuity tool a unit uses, whether its output is binding, and who can override it is asking the most informative staffing question available.
The alternative regimes
Four states set minimum nurse-to-patient ratios in law, and a larger group requires hospitals to run staffing committees with substantial direct-care nurse membership to set unit plans instead. The committee model is more flexible and much weaker, since a committee can be outvoted while a statutory maximum cannot. Both are described under nurse staffing and patient outcomes.
Long-term care sits outside all of this. Federal rules for nursing homes set no minimum hours per resident day at all. See nursing home staffing requirements.
Related
- Nurse staffing and patient outcomes
- Nurse strikes and bargaining
- Nursing home staffing requirements
- Twelve-hour shifts
- Nurse burnout
Sources
- Principles for nurse staffing. American Nurses Association. Accessed September 9, 2026.
- Nurse staffing measures. American Nurses Association, on registered nurse hours and total nursing care hours per patient day. Accessed September 9, 2026.
- Workforce management, patient classification and acuity systems, and the procurement process. American Nurses Association. Accessed September 9, 2026.
- Nurse staffing and inpatient hospital mortality. Needleman J, Buerhaus P, Pankratz VS, et al. New England Journal of Medicine 2011;364(11):1037-1045, for exposure to shifts below a unit's own target. Accessed September 9, 2026.
- 46,000 walked out in six weeks: what the 2026 nurse strikes actually won. NurseRecruiter, 3 March 2026, on the first contract limits covering staffing algorithms. Accessed September 9, 2026.