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.
Reviewed against primary sources on July 19, 2026 by the Soon operations research team
The evidence in one line
In the landmark hospital nurse survey, each additional patient in a nurse's workload was associated with 23% higher odds of burnout (OR 1.23, 95% CI 1.13-1.34) and 15% higher odds of job dissatisfaction (OR 1.15, 95% CI 1.07-1.25), adjusted for nurse and hospital characteristics (Aiken et al., 2002). The design is cross-sectional, so it establishes that heavier assignments and worse nurse outcomes appear together, not that one produces the other.
The workforce association is the largest one in this literature
Aiken et al. (2002) surveyed hospital nurses and linked their reported workloads to hospital data. After adjusting for nurse and hospital characteristics, each additional patient per nurse was associated with 23% higher odds of burnout (OR 1.23, 95% CI 1.13-1.34) and 15% higher odds of job dissatisfaction (OR 1.15, 95% CI 1.07-1.25).
The same study, using the same exposure and the same adjustment set, produced a mortality odds ratio of 1.07. Set the three coefficients side by side and the workload signal is roughly three times the size on the odds scale where it touches the workforce compared with where it touches patient survival.
Two cautions travel with those figures. They are odds ratios, not risk differences, so 23% describes a change in odds and cannot be read as a headcount of burned-out nurses. And the design is cross-sectional, so the authors do not claim that one of these things produced the other, and neither should you.
Where workloads differed across states, nurse outcomes differed with them
Aiken et al. (2010) compared nurses in three states. California nurses cared for 4.1 patients each overall, versus 5.4 in both New Jersey and Pennsylvania. On medical-surgical units the gap was wider: 4.8 patients in California against 6.8 in New Jersey and 6.5 in Pennsylvania.
Nurse outcomes lined up with those workloads. High burnout was reported by 29% of California nurses, against 34% in New Jersey and 36% in Pennsylvania. Job dissatisfaction was 20% in California, against 26% in New Jersey and 29% in Pennsylvania. The ordering is consistent across both measures and both comparison states.
What this comparison is not matters as much as what it is. It is descriptive, a snapshot of three states at one time. It is not a pre and post design, and it is not a difference-in-differences estimate. States differ in labor markets, hospital ownership mix, patient populations, and much else, and none of that is held constant here. The comparison is best read as consistent with the within-study association reported by Aiken et al. (2002), not as independent confirmation of a mechanism.
A retention argument that does not need the mortality question resolved
The patient-outcome side of this literature is genuinely contested, and the strength of the workforce evidence does not depend on how that argument ends. If you are an operations or nursing leader, the practical consequence is that you can build a staffing case on the association that is largest on the odds scale and closest to the decisions you actually control.
Assignment size is a scheduling variable. Burnout and job dissatisfaction are things you can survey on your own units this quarter. The distance between the published exposure and your operational lever is short, which is not true of a mortality endpoint measured within 30 days of admission that sits behind case mix, acuity, and a dozen clinical processes you do not set.
Be precise about where the evidence stops. Aiken et al. (2002) and Aiken et al. (2010) measured burnout and job dissatisfaction. They did not measure turnover, vacancy rate, agency spend, or time to fill. If your case runs from dissatisfaction to resignation to replacement cost, say plainly that the later steps come from your own data or from other literature, not from these two studies.
What this evidence cannot tell you
It cannot tell you which way the arrow points. A cross-sectional design cannot separate the possibility that heavy assignments wear nurses down from the possibility that hospitals with strained, dissatisfied workforces end up short-staffed, or that some third factor such as chronic underinvestment drives both. All three readings fit the data equally well.
It cannot rule out unmeasured confounding, and for these outcomes the plausible candidates are specific. Burnout and job dissatisfaction respond to the quality of the immediate nurse manager, how much mandatory overtime a unit runs, whether support staff absorb non-nursing work, how far in advance the schedule is published, and how much say nurses have in unit decisions. None of that was in the adjustment set, and any of it could sit behind both a heavy assignment and a discouraged workforce.
It cannot give you an expected result. No study here tested a staffing change and observed what happened to burnout. Aiken et al. (2010) documented that workloads and nurse outcomes differed across states at one point in time, which is a correlation across settings rather than the result of a policy. If you change assignments on your units, treat the outcome as an open empirical question and collect the before and after numbers yourself.
What this means for your schedule
- Track assignment size at the unit and shift level, since a hospital-wide monthly average hides the assignments nurses actually carry. Aiken et al. (2002) measured a hospital-level mean patient-to-nurse ratio, so unit and shift tracking is a deliberate refinement beyond what the study measured, not a replication of it.
- State the burnout and dissatisfaction figures on the odds scale when you present them, since 23% higher odds of burnout is not the same statement as 23% more burned-out nurses.
- Measure burnout and job dissatisfaction on your own units alongside workload before you change anything, so you have a baseline that belongs to you rather than a borrowed odds ratio.
- Refuse to promise a specific burnout reduction from a specific staffing change, because no study in this evidence base tested that intervention.
- Use the medical-surgical figures from Aiken et al. (2010) rather than the all-unit averages when your argument concerns medical-surgical units, since the gaps between states were wider there.
The business case
The staffing question can be argued on retention grounds without waiting for the patient-outcome debate to settle. Hospitals whose nurses carried heavier average assignments were also the hospitals where more nurses reported burnout and job dissatisfaction (Aiken et al., 2002), and those are the measures that sit closest to vacancy, agency spend, and orientation cost.
The honest version of the pitch is this: heavier assignments and worse nurse outcomes travel together across every setting studied, the direction of that relationship is not established by these designs, and no study here priced the tradeoff. Fund a measured change on your own units and hold it to your own before-and-after numbers rather than to a published odds ratio.
Frequently asked questions
- How much does each extra patient per nurse matter for burnout?
- In Aiken et al. (2002), each additional patient per nurse was associated with 23% higher odds of burnout, with a 95% confidence interval of 1.13 to 1.34 around the odds ratio of 1.23. The estimate is adjusted for nurse and hospital characteristics. Read it as an association from a cross-sectional design, not as a forecast of what burnout would do if you changed an assignment.
- Why is the burnout association larger than the mortality association?
- Because both odds ratios come from the same study. In Aiken et al. (2002), each additional patient per nurse was associated with OR 1.23 for burnout and OR 1.15 for job dissatisfaction, against OR 1.07 for death within 30 days of admission, all from one exposure and one adjustment set. The gap is therefore not an artifact of different populations or different models. A larger odds ratio is not a graver outcome, and the two should not be traded off against each other.
- What did the California comparison actually show?
- Aiken et al. (2010) compared three states and found California nurses cared for 4.1 patients each overall versus 5.4 in both New Jersey and Pennsylvania, and 4.8 on medical-surgical units versus 6.8 in New Jersey and 6.5 in Pennsylvania. Nurse outcomes tracked those workloads: high burnout was 29% in California versus 34% in New Jersey and 36% in Pennsylvania, and job dissatisfaction was 20% versus 26% and 29%. This is a descriptive cross-state comparison, not a pre and post or difference-in-differences causal estimate.
- Do these studies show that understaffing causes nurses to quit?
- No. Aiken et al. (2002) measured burnout and job dissatisfaction on a nurse survey, and Aiken et al. (2010) reported the share of nurses in each state with high burnout and with job dissatisfaction. Neither result is a measurement of who left, and neither study followed nurses over time to see whether dissatisfied nurses resigned. Treat the step from dissatisfaction to actual departure as a separate question that these two studies do not answer.
Sources
Every figure on this page is drawn from a cited primary source and checked against the original publication.
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)
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 →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.
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 →Staffing coverage math
Related resource
Open →