Does nurse staffing affect patient outcomes?
A four-article pillar on nurse workload and patient outcomes: what the landmark studies measured, what the California mandate research settled, and where the causal line actually sits.
Reviewed against primary sources on July 19, 2026 by the Soon operations research team
The evidence in one line
The strongest thing this evidence base offers is convergence, not proof. Two large cross-sectional studies, one in Pennsylvania and one across 300 hospitals in nine European countries twelve years later, produced nearly the same adjusted coefficient for nurse workload against death within 30 days of admission: a coefficient near 1.07 on the odds scale in both Aiken et al. (2014) and Aiken et al. (2002). Both are associations measured on the odds scale, and no study in this literature has established that changing a hospital's staffing changes its mortality.
What the evidence establishes
Start with the convergence rather than with any single number. Aiken et al. (2002) surveyed hospitals in Pennsylvania. Aiken et al. (2014) covered 422,730 surgical patients aged 50 and over in 300 hospitals across nine European countries. Different patients, different health systems, twelve years apart, and the two workload coefficients for death within 30 days of admission landed within a fraction of a percentage point of each other on the odds scale. The mortality article gives both figures, the scale they sit on, and the model they came from.
A third study asked whether the pattern held in states with sharply different staffing environments. Aiken et al. (2010) compared California, which had a legislated ratio floor, with New Jersey and Pennsylvania, which did not. All three state estimates pointed the same direction, with confidence intervals that overlap heavily. The deep article on the mortality question lays out the three state-specific figures and what that overlap does and does not license you to say.
Replication across independent samples is worth something. It makes it less likely that the pattern is an artifact of one hospital system, one case mix, or one measurement approach. It is not a causal demonstration. All three studies are cross-sectional, and their authors decline to claim cause. The mortality article works through the alternative explanations that fit the same data.
Where the association is largest: the nurses
The patient-mortality numbers get the headlines. The nurse-side numbers in the same data are considerably larger. In Aiken et al. (2002), the nurse-side associations with burnout and job dissatisfaction were the largest in the study, and the workforce article gives both figures.
Set those beside the mortality association from the same study and the workforce side is clearly the larger of the two. For an operations leader it is also the part of the evidence base with the shortest chain of inference. Burnout and dissatisfaction were reported by the nurses who carried the workload, in the same survey that recorded the workload, so exposure and outcome came from one source rather than being joined across separate datasets. Exposure and outcome came from the same survey respondents rather than being joined across separate datasets.
It is also the part you are most likely to be able to observe in your own unit without a research team. Turnover, sick calls, and exit interviews are already in front of you. The workforce article takes this further, including what the burnout and dissatisfaction measures actually captured and what they left out.
The causal gap, stated plainly
The obvious next question is whether mandating better ratios would move patient outcomes. The available quasi-experimental evidence answers only the first half of that. McHugh et al. (2012) found that California's AB 394 was followed by a real drop in patients per licensed nurse close to one patient per nurse, which is the same unit the association studies are denominated in. The staffing side moved. The outcome side was never connected to it, by that study or by any other, and the mandate article works through why the design cannot carry that connection.
So the honest position has two halves, and both matter. Nobody has established that mandated ratios save lives. Nobody has established that they failed to help patients either.
Projections turn up often in this literature: a model contrasting a well-staffed hospital profile with a poorly staffed one, per-1000 absolute risk figures, counts of deaths that would differ under alternative staffing. Every one of those is an extrapolation from correlational coefficients rather than an observed outcome. Reproduce them where they help, and label them as model output every time.
Where to go from here
Three deeper articles sit under this hub. The first works through the core mortality association in detail: what each of the three landmark studies measured, on what scale, with what adjustments, and what an odds ratio near 1.07 does and does not tell you about an individual patient.
The second takes the mandate question directly, including what the California evidence supports about staffing levels, why the outcome question remains open, and how to read the projections that get quoted as though they were findings.
The third covers the workforce side, where the association is strongest, where the exposure and the outcome came from the same respondents, and where an operations leader has the clearest line of sight to something they already control.
What this means for your schedule
- Describe these findings as associations, not effects, when you present them to a board or a unit council, because every mortality estimate here comes from a cross-sectional design whose authors disclaim causality.
- Match the claim you want to make to the study that actually supports it: patient-outcome estimates come from the cross-sectional Aiken analyses, a change in staffing levels under a mandate comes from McHugh et al. (2012), and nothing in this evidence base joins the two.
- Weight the nurse-side evidence heavily in your own planning, since burnout and job dissatisfaction were the largest and most directly measured associations in Aiken et al. (2002).
- Label any per-1000 risk figure or projected death count as a model extrapolation from correlational coefficients whenever you reproduce one.
- Refuse both confident positions on mandated ratios, because the California evidence covers what happened to staffing levels and says nothing verified about what happened to patients.
The business case
The staffing signal in this literature is unusually consistent: two independent samples a decade and an ocean apart produced nearly the same adjusted coefficient for 30-day mortality per patient added to the average nurse workload, and the three states in Aiken et al. (2010) pointed the same way.
The associations with nurse burnout and job dissatisfaction were larger still, which puts workload squarely in the retention conversation where the cost consequences are immediate and measurable.
What the evidence does not yet support is a return-on-investment claim tied to patient survival, so build the case on workforce stability and treat the mortality association as context rather than as a projected saving.
Frequently asked questions
- Do more nurses cause fewer patient deaths?
- No study in this evidence base establishes that. Aiken et al. (2002) and Aiken et al. (2014) both found nurse workload associated with higher adjusted odds of a surgical patient dying within 30 days of admission, but both used cross-sectional designs in which staffing and outcomes were observed at the same time. The authors do not claim to have identified a causal relationship, and neither should anyone citing them.
- Are the European and US findings measuring the same thing?
- Closely enough that the agreement is meaningful. Aiken et al. (2002) covered Pennsylvania hospitals; Aiken et al. (2014) covered 422,730 surgical patients aged 50 and over across nine European countries. Both used a hospital-level nurse workload measure against death within 30 days of admission, and the adjusted coefficients came out near each other. One caution travels with the 2014 result: its partly adjusted model returned OR 1.005 (p=0.816), a null. The mortality article works through what changed between the two models.
- Why does a page about patient outcomes carry a policy finding at all?
- Because the policy research is the only place in this literature where staffing actually changed and was then measured, which is the design outcome questions need. McHugh et al. (2012) showed that a legislated floor moved real workloads in California hospitals. It stops short of what happened to patients, so the policy evidence marks the boundary of the outcome question rather than answering it. The mandate article covers the design and its limits.
- Is the association stronger for patients or for nurses?
- For nurses, by a substantial margin. In Aiken et al. (2002), each additional patient per nurse was associated with OR 1.23 (95% CI 1.13-1.34) for nurse burnout and OR 1.15 (95% CI 1.07-1.25) for job dissatisfaction, against OR 1.07 (95% CI 1.03-1.12) for death within 30 days of admission in the same study. The workforce associations are also more directly measured, since both the exposure and the outcome came from the nurses themselves.
Sources
Every figure on this page is drawn from a cited primary source and checked against the original publication.
Aiken, L. H., Sloane, D. M., Bruyneel, L., Van den Heede, K., Griffiths, P., et al. (2014). Nurse staffing and education and hospital mortality in nine European countries: a retrospective observational study. The Lancet, 383(9931), 1824โ1830. https://pmc.ncbi.nlm.nih.gov/articles/PMC4035380/
Retrospective cross-sectional observational study (422,730 surgical patients, 300 hospitals, 9 countries)
Aiken, L. H., Clarke, S. P., Sloane, D. M., Sochalski, J., & Silber, J. H. (2002). Hospital Nurse Staffing and Patient Mortality, Nurse Burnout, and Job Dissatisfaction. JAMA, 288(16), 1987โ1993. https://doi.org/10.1001/jama.288.16.1987
Cross-sectional analysis of survey and discharge data (Pennsylvania hospitals)
Aiken, L. H., Sloane, D. M., Cimiotti, J. P., Clarke, S. P., Flynn, L., Seago, J. A., Spetz, J., & Smith, H. L. (2010). Implications of the California Nurse Staffing Mandate for Other States. Health Services Research, 45(4), 904โ921. https://doi.org/10.1111/j.1475-6773.2010.01114.x
Descriptive cross-state comparison (California, New Jersey, Pennsylvania)
McHugh, M. D., Brooks Carthon, M., Sloane, D. M., Wu, E., Kelly, L., & Aiken, L. H. (2012). Impact of Nurse Staffing Mandates on Safety-Net Hospitals: Lessons from California. The Milbank Quarterly, 90(1), 160โ186. https://pmc.ncbi.nlm.nih.gov/articles/PMC3371663/
Within-California pre/post fixed-effects time series (173 hospitals, 1998-2007)
None of the studies cited here evaluated Soon.They examine scheduling practices, shift patterns, and working hours as studied by independent researchers, so their findings describe what those practices are associated with, not what any particular software produces.
This article summarizes published research for scheduling and operations decisions. It is not medical advice. Individual health questions belong with a qualified clinician.
Keep reading
What does the research actually show about nurse staffing and patient mortality?
Three landmark studies found the same association between nurse workload and death within 30 days of admission, on the odds scale, in designs that cannot establish cause.
Read →Do mandated nurse-to-patient ratios save lives?
A legislated ratio floor was followed by a real drop in how many patients each nurse carried, but whether it changed patient survival has not been established, and the two numbers most often quoted as proof are model projections.
Read →What does understaffing do to nurses?
The workload association with burnout and job dissatisfaction is the strongest signal in the nurse staffing literature, and it stands independently of the contested mortality question.
Read →Healthcare scheduling
Related resource
Open →