Beyond Nurse Staffing Ratios: What Does Nursing Actually Produce?

Part of Nurse.org’s Nursing AI Watch, our ongoing investigation into how artificial intelligence is reshaping nursing practice. This is part four of The Nursing Value Project, nurse and clinical AI strategist Mark Smith’s series on how healthcare measures and values nursing. Parts one, two, and three asked how software got an economic language before nursing, why decades of measurement research never became infrastructure, and what nurse leaders can do about it. This installment turns to the newest research: what nursing actually produces, and what it would take to prove it.

We know how many nurses work in a hospital, what they cost, and how many patients they are assigned. But do we know enough about the nursing care each patient receives, or the difference that care makes?

Imagine two registered nurses working on the same medical-surgical unit. Each is assigned four patients.

One is caring for patients whose conditions are relatively stable. Several are preparing for discharge. Their needs include medication administration, monitoring, education, and coordination of follow-up care.

The other has a patient whose condition is deteriorating, another with significant cognitive impairment and a high risk of falling, and two others requiring frequent reassessment.

Both nurses have the same staffing ratio.

Their workloads are not equivalent.

Neither are the clinical demands placed on them, the time required to meet their patients’ needs, or the consequences of care being delayed or missed.

This difference illustrates a longstanding problem in healthcare, one this series has traced from its history to a playbook for nurse leaders. Hospitals can calculate nursing labor costs with considerable precision, yet often struggle to connect those expenditures to the care individual patients receive and the outcomes that follow.

If nursing is to be valued as more than an institutional expense, that gap must be addressed.

Want to see more Nurse.org articles in your Google results? Add us as a preferred source.

Staffing ratios matter.

Decades of research have established an important relationship between registered nurse staffing and patient safety.

A 2022 systematic review by Dall’Ora and colleagues examined 27 longitudinal studies of nurse staffing and found evidence consistent with lower patient mortality when registered nurse staffing was higher.

The review also identified methodological limitations. Only three studies were judged to have a low overall risk of bias, making precise estimates of the effect of staffing changes difficult.

These limitations do not diminish the importance of safe staffing. They demonstrate why establishing the size of a clinical benefit requires more than observing an association.

Staffing ratios provide an essential measure of available nursing capacity and patient workload. They are important safeguards for patients and nurses alike.

But a ratio does not tell us how much nursing care a patient needs, what care was delivered, or whether necessary care was delayed or omitted.

Patient acuity, clinical instability, admissions, discharges, interruptions, and available support staff can substantially change the demands of a shift without changing the ratio.

Meeting a staffing standard is therefore not the same as demonstrating that every patient’s nursing needs have been met.

To understand nursing’s contribution, we must examine what happens after nurses are assigned to patients.

In 2009, nursing researcher John M. Welton and colleagues published a study investigating whether nurse–patient assignment records could provide more precise estimates of nursing intensity and direct nursing costs.

Using 11,582 patient-days at a single university hospital, the researchers compared assignment-based estimates with measured nursing intensity.

Their approach explained 77.2% of the variation in daytime nursing intensity and 82.4% at night, outperforming estimates based on average nursing hours at the unit or hospital level.

The findings demonstrated something important: knowing which nurses are assigned to which patients provides information that aggregate staffing averages cannot.

But an assignment is not a complete record of nursing care.

It establishes a connection between a patient and the nursing resources allocated to that patient’s care. It does not tell us precisely how much time was spent with the patient, every intervention performed, or the effectiveness of that care.

Patients may receive care from several nurses during a shift. Assignments change. Nurses assist colleagues during emergencies, procedures, and transfers.

Some of nursing’s most consequential work also involves monitoring, anticipating deterioration, and making clinical judgments that may not appear as distinct activities in the electronic health record.

The assignment is therefore a useful starting point for estimating nursing resource use.

It is not, by itself, a measure of nursing’s clinical contribution.

If assignment records help establish where nursing resources are directed, clinical documentation offers information about the care provided.

Hospitals routinely record medication administration, assessments, monitoring, interventions, patient education, and care coordination.

These records are useful, but they cannot fully explain nursing practice.

Consider a nurse who recognizes a subtle change in a patient’s condition and promptly escalates the concern.

The response may involve a physician, pharmacist, respiratory therapist, and several nurses. The patient may stabilize without requiring transfer to an intensive care unit.

The nurse’s recognition and escalation may have been essential. But the outcome reflects the work of multiple clinicians, and we cannot assume that an adverse event would otherwise have occurred.

That is one reason counting completed nursing tasks cannot establish nursing value.

There is also the matter of care that should have occurred but did not.

A 2025 systematic review by He and colleagues examined 17 studies investigating the relationship between missed nursing care and patient outcomes.

Missed monitoring, basic care, treatment, and symptom management were associated with adverse outcomes, including falls, urinary tract and bloodstream infections, readmissions, and mortality. Missed communication and basic care were also associated with lower patient satisfaction.

Most studies were cross-sectional and relied on nurse or patient reports, limiting the strength of causal conclusions.

Nevertheless, the findings reveal an important weakness in conventional productivity measures.

A record showing that a nurse completed numerous activities tells us little about the care that was needed but left unfinished.

Nor does a shift with relatively few documented interventions necessarily represent less nursing work. Continuous assessment, surveillance, readiness to intervene, and the maintenance of clinical stability are important parts of nursing practice.

A credible measurement system must account for patient needs, nursing care delivered, and unmet care needs.

Otherwise, we risk measuring documentation rather than care.

>>Listen to The Latest Nurse News Podcast

For decades, routine hospital nursing expenses have largely been incorporated into broader inpatient accounting and charging structures rather than separately attributed to individual patients.

Recent research demonstrates that more precise measurement is possible.

In 2026, Seo Yoon Lee and colleagues published a study estimating patient-level nursing costs using linked patient and nursing data.

The researchers reported a mean daily nursing cost of $586.23 per patient, representing 25.6% of the daily room charge.

Patient acuity, Medicare Severity Diagnosis-Related Groups, nursing experience, and patient load were significant predictors of nursing costs.

The study is important because it brings nursing cost estimation closer to the individual patients receiving care.

It also illustrates a distinction that is frequently overlooked in discussions of nursing’s financial value.

The cost of delivering nursing care is not the same as the financial value generated by that care.

The reported $586.23 represents an estimated daily nursing cost. It is not reimbursement received for nursing services. Nor does the comparison with room charges establish how much revenue nursing generated.

Hospital charges, operating costs, reimbursement, and economic value are different measures.

Under Medicare’s traditional inpatient prospective payment system, hospitals generally receive predetermined payments for inpatient cases, rather than separate payments for each routine bedside nursing service. Nursing costs are ordinarily included within the hospital’s broader inpatient payment structure.

That does not make nursing financially irrelevant.

Nursing can influence complications, resource utilization, length of stay, and the cost of delivering care.

But measuring what nursing costs does not automatically tell us what it saves, prevents, or contributes financially.

Patient-level costing provides one piece of the answer.

Another study published in 2026 provides an important example of what more precise nursing measurement can accomplish.

David Ryamukuru, Olga Yakusheva, and colleagues examined associations between specialty nursing certification and patient outcomes using data linking 42,078 patients with 2,683 nurses across four hospitals.

Rather than relying solely on hospital-wide measures of nursing qualifications, the researchers examined the proportion of care individual patients received from specialty-certified registered nurses.

Greater exposure to specialty-certified nurses was associated with better patient outcomes after adjustment for measured patient and nurse characteristics, including educational preparation.

The analysis is significant because it moves beyond asking whether hospitals employing more certified nurses have better outcomes.

It begins examining what happens to patients who receive care from those nurses.

That distinction opens important possibilities for studying nursing experience, assignment patterns, continuity, clinical expertise, and patient outcomes.

It also introduces methodological challenges.

A patient’s deteriorating condition may lead to more nursing attention, rather than the additional nursing attention causing deterioration. Patients receiving more intensive care may have worse outcomes because they were sicker to begin with.

Researchers must therefore account for patient risk, changing clinical conditions, the timing of nursing care, and contributions from other clinicians.

The certification findings are observational. They do not establish that certification itself caused better outcomes or that investments in certification generated a positive financial return.

What they demonstrate is the potential of linking nursing characteristics and patient exposure to individual outcomes.

That is an important step toward better measurement, but it is not yet a complete model of nursing value.

Consider a hypothetical hospital introducing a nursing intervention intended to improve surveillance and early recognition of patient deterioration.

The organization measures patient acuity, staffing, nurse–patient assignments, completed assessments, missed care, clinical escalation, intensive care transfers, length of stay, and nursing labor costs.

After implementation, fewer patients experience unplanned transfers to intensive care.

That is encouraging.

But before attributing the improvement to the nursing intervention, the hospital would need to establish what likely would have occurred without it.

Were patients comparable? Did their underlying risks change? Were other clinical programs introduced at the same time? Did documentation or transfer practices change?

A credible analysis needs a defensible comparison, appropriate adjustment, and evidence that the nursing intervention preceded the measured outcome.

Even if those requirements are met, demonstrating clinical benefit does not automatically establish financial return.

Suppose the intervention reduces hospital length of stay.

Shorter stays may reduce medication, supply, and other variable expenses. They may also free beds for additional patients.

But they do not necessarily reduce payroll, facility expenses, or other costs that remain fixed in the short term.

The hospital may realize financial benefit if expenses actually decline, if the available capacity is used productively, or if reimbursement and resource use change favorably.

It must also account for the costs of implementing and maintaining the nursing intervention.

Without that reconciliation, the organization may have demonstrated improved care and a potential economic benefit, but not an actual financial return.

This distinction is critical.

A frequently cited staffing study illustrates the problem.

In 2021, Karen Lasater and colleagues examined staffing and outcomes in Illinois hospitals.

Using observational findings, they estimated that staffing medical-surgical units at a four-patient-per-nurse ratio during the study period could have prevented approximately 1,595 deaths among Medicare patients.

They also projected more than $117 million in hospital cost savings associated with shorter lengths of stay.

Those projections were based on statistical estimates, including the use of cost-to-charge ratios to estimate the costs associated with fewer hospital days.

They were not observed financial savings following implementation of the proposed staffing ratios.

The study provides important evidence for staffing policy. But the projected savings should not be presented as money hospitals actually recovered or expenses they demonstrably eliminated.

Nor does the projection independently establish the net financial return after accounting for the full cost of changing staffing levels.

This is the standard that nursing economic research should apply consistently.

Claims of financial return require three elements:

1. A credible counterfactual: What would likely have happened without the intervention?
2. Transparent costing: Which expenses, resource changes, and financial assumptions were included?
3. Demonstrated financial realization: How did the measured benefit affect actual expenditures, payments, or productive capacity?

An intervention may have substantial clinical value without meeting all three requirements.

That is not an argument against the intervention.

Safe staffing, timely care, and patient protection are obligations regardless of whether they produce a favorable return on investment.

Economic evaluation should help organizations understand the resources required to meet those obligations, not determine whether patients deserve essential care.

Nursing researchers are proposing several ways to address these problems.

In 2024, Welton and Robert Longyear proposed alternative billing and reimbursement approaches, including separating nursing care from daily facility charges and expanding procedure-based mechanisms for nursing services.

In 2026, Emilie A. Maxie introduced the Nursing Value Index, a conceptual framework organized around preventing adverse outcomes, reducing their severity, and maintaining clinical stability.

These proposals address important aspects of nursing valuation, although conceptual approaches still require empirical testing before their attribution methods can be treated as validated.

For hospitals, a practical starting point is to connect information that already exists across staffing, clinical, and financial systems.

Nurse–patient assignments can help identify how nursing resources are allocated. Patient acuity and clinical assessments can describe care requirements. Clinical records and missed-care measures can help characterize care delivered and omitted. Outcome and financial data can support subsequent evaluation.

But assembling those records is not necessarily straightforward.

In 2026, Lee and colleagues described difficulties constructing a Nursing Value Model from existing datasets because identifiers for nurses and patients were not consistently compatible across systems.

Better data integration, consistent identification, and reliable measurement are necessary before these approaches can be widely implemented.

Hospitals must also avoid turning imperfect data into simplistic productivity rankings.

A nurse caring for clinically complex patients should not be penalized for treating patients with greater baseline risks. Outcomes produced through interdisciplinary care cannot fairly be assigned to one nurse without a defensible attribution method.

The appropriate unit of analysis will often be the patient episode, nursing team, or clinical intervention rather than individual nurse productivity.

The goal is to understand nursing’s contribution, not simply extract more completed tasks from every nursing hour.

One additional dimension deserves investigation: the therapeutic relationship between a nurse and patient.

Trust, continuity, communication, and familiarity with an individual’s condition may influence how effectively nursing care is delivered and received.

Those potential effects deserve their own measurement and validation. They should not be assumed or assigned financial values without evidence.

Hospitals have spent decades refining how they account for nursing labor.

They can report staffing hours, payroll expenses, productivity targets, and departmental budgets.

What remains much more difficult is connecting those resources to the care individual patients receive, the outcomes that follow, and the economic consequences that can be credibly attributed to nursing.

The research is beginning to address that gap.

Nurse–patient assignments provide a foundation for estimating patient-level resource use. Contemporary costing methods are making nursing expenditures more visible. Assignment-linked research is helping investigators examine relationships between nursing characteristics and patient outcomes.

None of these approaches, alone, constitutes a complete valuation system.

A defensible approach must distinguish staffing capacity from actual care, nursing cost from clinical contribution, and modeled economic benefits from realized financial returns.

Staffing ratios remain indispensable. Nurses have fought to establish and defend them, and they help create the conditions under which safe nursing care can be delivered.

But they cannot tell us everything nursing produces.

To answer that question, healthcare must move beyond counting nurses and begin measuring the care patients need, the care they receive, and the difference nursing makes, with enough rigor to distinguish what the evidence demonstrates from what we merely hope or assume.

We know what hospitals spend on nursing. The next challenge is determining, with the same discipline, what nursing contributes.

The Nursing Value Project (complete series):

🤔 Think about your last shift. What did you prevent, catch, or hold together that no staffing ratio or productivity report will ever show? Share your experience in the comments below.

Nurse.org Analysis