What the Challenger Disaster and Nurse Staffing Have in Common
A root cause analysis of normalization of deviance in nursing finance
The morning of January 28, 1986, engineers at Morton Thiokol tried to stop the Challenger launch. They had documented their concerns in writing months earlier. The O-rings sealing the solid rocket boosters were not designed to function in the overnight temperatures that had fallen across Cape Canaveral. They had evidence. They had charts. They said: do not launch.
NASA managers pushed back. Under pressure, Thiokol management reversed course and authorized the launch. Seventy-three seconds after liftoff, Challenger broke apart. Seven crew members died.
Sociologist Diane Vaughan spent years studying what happened. Her 1996 book, The Challenger Launch Decision, gave a name to the mechanism she found: normalization of deviance (Vaughan, 1996). When an organization encounters a rule deviation or warning signal that does not immediately produce a catastrophe, it begins to treat the deviation as acceptable. The warning stops being a warning. It becomes the standard. Over time, the deviance becomes normal, and the people inside the organization cannot see it anymore.
The O-rings had eroded on previous flights. Each time, the mission had succeeded anyway. Engineers raised concerns. Managers weighed the concerns against the launch schedule, the cost pressures, the political visibility of the shuttle program, and they authorized flight. The system worked, again and again, until it did not.
Challenger is often invoked as a communication failure, proof that the engineers simply presented their data badly. That lesson is too small. Better slides would not have undone the years in which O-ring erosion was observed, tolerated, and reclassified as an acceptable risk. The presentation problem, if there was one, sat downstream of a deviation the organization had already stopped seeing. The same holds in nursing finance. The formula error is not a message poorly explained. It is a deviation the profession normalized.
Normalization of deviance is not solely a NASA problem. Vaughan’s framework describes what happens inside any organization when warning signals are consistently managed away rather than investigated. The pattern requires only two conditions: a deviation that does not produce an immediate, visible catastrophe, and an institutional structure that creates pressure to proceed regardless.
Both conditions are fully present in nursing finance.
The Error
For sixty-six years, many nurse staffing budgets have been built on a formula that contains a mathematical error. The error produces a budget that is short by two to six percent before the fiscal year begins, before a single staffing decision is made, before a single shift is built.
William J. Ward, Jr., MBA, Associate Professor of Health Finance and Management at the Johns Hopkins Bloomberg School of Public Health, identified the error directly in his healthcare budgeting textbook:
“Even those skilled in budget calculations often make the mistake of saying, ‘Let’s add 15% for nonproductive time.’ But this approach is clearly incorrect and will always result in a budget that lacks sufficient staff.” (Ward, 2016, pp. 129-130)
The proof is short.
Start with a unit that needs 80 productive FTEs to deliver direct patient care. Nonproductive time, which includes paid time off, holidays, sick time, and required education, represents 20% of total paid time. Twenty percent is higher than current US rates and lower than some international benchmarks. It is used here because it produces round numbers that make the arithmetic transparent and the error easy to identify. The remaining 80% is available for care.
The numbers are given: 80 productive FTEs, 20 relief FTEs, 100 total FTEs.
80 + 20 = 100.
The question any nurse leader faces when building a staffing budget is: given 80 productive FTEs required, how many total FTEs must be budgeted?
The standard approach in use across the nursing literature and CE offerings applies the nonproductive rate to the productive FTE count to determine the relief staff to add:
80 × 0.20 = 16 relief FTEs
80 + 16 = 96 total FTEs
The budget is short by four FTEs. We started with 100. The formula produces 96.
Where did the 4 FTEs go?
The formula treats the nonproductive rate as a percentage of the productive workforce. The rate is a percentage of the total workforce. The 16 relief FTEs are subject to the same 20% nonproductive rate as every other position in the budget. To find the correct total number of relief FTEs, apply the same division to the 16:
16 ÷ 0.80 = 20 relief FTEs
20 − 16 = 4 FTEs recovered
The four recovered FTEs cover the nonproductive time of the relief staff themselves. The correct total:
80 + 20 = 100 total FTEs
Or, applying the division directly to the productive FTE requirement:
80 ÷ 0.80 = 100 total FTEs
The budget is whole.
Ward puts it plainly: “Avoid the mistake of merely adding 15% of nonproductive time back to the 17 workers. As demonstrated in Table 6.14, doing so will short change the FTE count and leave a manager short staffed even before the fiscal year begins.” (Ward, 2016, p. 129)
Sixty-Six Years of Evidence
The error did not start with any individual. It was not introduced by one careless author or one bad textbook. Other valid calculation approaches for nursing workforce planning exist; this analysis addresses the one that appears most consistently across the nursing literature, professional education, and practice settings. Tracing it requires following a line that runs through six decades, multiple countries, and the professional organizations responsible for nursing education itself.
1960 | United States
Sister Mary Laura Gunn’s 1960 monograph, A Method for Developing a Master Staffing Plan for the Nursing Service Department, published by the Catholic Hospital Association, contains the earliest known instance of the calculation error (Gunn, 1960). Gunn was a graduate of a hospital administration program at Saint Louis University. Her diagnosis of the problem was correct: nursing shortages were in many instances “the result of faulty planning by the administrators of the nursing service department.” Her formula was not. At the 6% nonproductive rate she used, the structural shortfall was approximately 0.36 FTEs per 100 budgeted staff. That amount disappears inside normal scheduling variance. At 6%, the formula appeared correct because it nearly was.
1979 + 2003 | United Kingdom
Keith Hurst’s systematic review of nursing workforce planning methods, commissioned by the UK Department of Health and published in 2003, documented and credited W.A. Telford as the originator of the additive formula in UK nursing, attributing it to a 1979 article in Hospital Health Services Review and a 1983 government report for North Birmingham Health Authority (Hurst, 2003). Hurst named Method 1 the Telford Method, but the additive formula error runs through all five workforce planning approaches in the review. Each method uses a different process to estimate care hours required; all five then feed that estimate into the same flawed conversion step. Where he shows the arithmetic explicitly, structural shortfalls reach approximately 5% at the 22% time-out rate then in use. The Department of Health adopted the review as NHS guidance. What Hurst transmitted as Telford’s method, the Department of Health adopted as policy.
1984 | United States: The Warning That Did Not Travel
Two years before the additive formula appeared in a major US nursing journal, the primary graduate textbook in nursing budgeting placed the correct formula on record and labeled the alternative wrong.
Budgeting Concepts for Nurse Managers, edited by Steven Finkler and published by W.B. Saunders, presented both approaches in Chapter 4. A footnote on page 73 stated: “Although the method described here is widely used, it is not theoretically correct” (Graf, 1984, p. 73). The footnote named the mechanism: staff hired to cover nonproductive time are themselves nonproductive some fraction of the time. It presented the division formula as the correct alternative and noted that the difference “can amount to a large amount of money when accumulated for all staffing needs for all nursing units.” The 1992 second edition carried the same acknowledgment. The correct formula was in the primary graduate text from 1984 forward. It did not stop what followed.
1987 | United States
Leah Strasen’s Key Business Skills for Nurse Managers, published by J.B. Lippincott in its Nursing Management Series, presented the additive formula without qualification (Strasen, 1987). The worked example at pages 136-137: 28.8 patient care FTEs × 1.15 (benefit time = 15% per FTE) = 33.1 total FTEs. The correct total at 15% nonproductive is 33.88. Strasen was a vice president of nursing at Henry Mayo Newhall Memorial Hospital in California, writing in a series for nursing leader professional development.
1999 | United States: Published by the Editor
The June 1999 issue of Nursing Management published “How to Develop a Unit Personnel Budget” (Brown, 1999). The formula: (ADC × NHPPD × 1.4 × 1.14) / 7.5 = total caregiver FTEs. The 1.14 multiplier applies a 14% nonproductive rate to the productive FTE count. The correct total at 14% nonproductive requires dividing by 0.86. The structural shortfall is approximately 2%.
The author, Barbara Brown, was the Clinical Editor of Nursing Management at the time of publication. The journal’s own editor published the error in the journal she edited. The formula Brown introduced, with the same 1.14 multiplier and additive structure, appears in a graduate nursing administration textbook published twenty-one years later.
2006 | International: The Error That Traveled
The International Council of Nurses published Safe Staffing Saves Lives as an International Nurses Day toolkit in 2006, designed for use across its member countries worldwide (International Council of Nurses, 2006). Annex 6, explicitly adapted from Hurst’s 2003 UK Department of Health review, reproduced the additive formula. The worked example: 455 hours × 1.22 (time-out factor) = 555.1 hours ÷ 37.5 hours per WTE = 14.8 WTEs required. The correct calculation: 455 ÷ 0.78 = 583.3 hours ÷ 37.5 = 15.56 WTEs. The structural shortfall at 22% nonproductive is approximately 4.9%.
The error traveled from a UK government-commissioned systematic review to a Geneva-based federation of national nurses’ associations with member organizations across the globe. The 22% nonproductive allowance it presented as an international benchmark became the figure a US book on nurse staffing carried across three successive editions as the basis for its 1.22 multiplier.
2018 | United States: Continuing Education
The May 2018 issue of Nursing Management published “Developing a Staffing Plan to Meet Inpatient Unit Needs,” carrying CE credit toward ANCC and AACN certification renewal (Hunt, 2018). The instructional method was the additive formula. Nurse leaders applying this formula would budget 2 to 4 fewer FTEs per 100 than their units required, depending on the nonproductive rate in use. By 2018, Ward’s correction had been in print for two years. The footnote in Graf and Finkler identifying the correct formula had been on record for thirty-four years.
2020 | United States
Management and Leadership for Nurse Administrators, eighth edition, published by Jones & Bartlett Learning, presents the staffing budget formula in Box 8-3 of Chapter 8:
(ADC × HPPD × 1.4 × 1.14) / 7.5
The formula is structurally identical to the one Brown published in Nursing Management in 1999: same 1.14 multiplier, same additive structure. The chapter author, Patricia Thomas, is also one of the book’s three editors. Sixty years after Gunn, the error appears unchanged in a graduate nursing administration text used to train the people who will teach the next generation of nurse leaders (Thomas & Roussel, 2020).
2026 | United States
By April 2026, the formula was still being taught to nurse leaders as continuing education. An ANCC-accredited CE workshop delivered to practicing nurse leaders this spring showed the following at Step 5 of its staffing plan build:
2,080 Annual hours for 1 FTE
281.7 Hours of non-productive time281.7 ÷ 2,080 = 13.5% lost time
47.86 FTE + 13.5% = 54.4 + 1.0 = 55.4 FTE needed
The 1.0 is a nurse manager FTE added separately. The core calculation, 47.86 FTE + 13.5%, is the additive formula, identical in structure to the one Gunn used in 1960. The error has not been corrected. It is being actively taught. That workshop was not isolated. Four additional nursing leadership workshops attended in the preceding five years each presented the same formula without correction.
The individuals and organizations documented here are not the problem. Each reproduced a methodology the profession had treated as authoritative for decades before they encountered it. The error was not visible at the nonproductive rates when the earliest sources were written. By the time rates had grown large enough to produce operationally meaningful shortfalls, the formula had accumulated the authority of textbooks, government guidance, international standards, and CE certification. The formula is the problem. The people who transmitted it were working within the professional infrastructure as the profession had defined it.
Why the Error Survived Correction
The question is not only why the error appeared. It is why, after sixty-six years and two published corrections, it is still being taught.
At 6%, the error is nearly invisible.
Gunn’s 1960 formula used a 6% absence allowance, and, at that rate, the additive formula produces a shortfall of roughly 0.36 FTEs per 100 budgeted staff. A hospital using a 6% nonproductive rate and the wrong formula would budget 99.6 total FTEs where the correct answer is 100. The gap disappears inside normal scheduling variance. No audit catches it. No manager notices it. The formula appears correct because at low rates, it nearly is.
Leave entitlements did not stay at 6%.
Over the following decades, vacation allowances increased. Sick leave expanded. Mandatory continuing education became a credentialing requirement. Orientation periods grew as patient care complexity increased. Each addition increased the percentage of time unavailable for direct care. By the time the 2000s sources were published, nonproductive rates of 13 to 24 percent were common across the global nursing finance literature. At 20%, the additive formula produces a shortfall of 4.0 FTEs per 100. At 24%, the shortfall is 6.1 FTEs per 100.
The formula did not change. The shortfall grew.
This is the mechanism that kept the error hidden through its most consequential decades. At the rate it was introduced, it was undetectable. By the time the nonproductive percentage had grown large enough to produce operationally meaningful understaffing, the formula had accumulated decades of institutional authority. Questioning it required questioning textbooks, credentialing programs, government frameworks, and the accumulated professional knowledge of every finance director who had trained nurse leaders on how to build a budget.
Two additional features of nursing finance prevented detection.
Budget variance analysis does not examine the formula. It measures actual spending against the budgeted FTE count. A unit consistently running short-staffed generates a productivity variance, a turnover flag, or a premium labor overage. None of those analyses look upstream at whether the budgeted FTE count was correctly derived. The error was invisible to the oversight mechanism because the oversight mechanism assumed the budget was correct.
No verification standard exists either. Any FTE calculation can be checked in one step: multiply the proposed total FTEs by the productive percentage and confirm the result equals the original productive FTE requirement. If the numbers do not match, the budget does not fund the care hours it claims to fund. That check does not appear in any of the more than forty published sources where the error has been documented, spanning six decades and multiple countries. It does not appear in the CE workshops where the error is taught. Without a verification step, no practitioner applying the formula has a built-in mechanism to detect the shortfall.
The Human Toll
Seven people died when Challenger broke apart on January 28, 1986. Their names are Christa McAuliffe, Gregory Jarvis, Judith Resnik, Francis Scobee, Ronald McNair, Ellison Onizuka, and Michael Smith. A presidential commission was convened within days. The cause was identified within months. The program was grounded. Procedures were changed.
No equivalent investigation has been launched for nursing finance.
Three decades of peer-reviewed research has established a direct relationship between nurse understaffing and patient mortality.
Aiken and colleagues, studying 168 Pennsylvania hospitals, found that each additional patient added to a nurse’s assignment was associated with a 7% increase in 30-day mortality and a 7% increase in failure to rescue (Aiken et al., 2002). Needleman and colleagues, studying a Magnet-designated hospital where actual mortality ran 39% below predicted rates, found that each shift in which registered nurse staffing fell 8 or more hours below target was associated with a 2% increase in patient mortality; 34.6% of patients were exposed to three or more such shifts during their admission (Needleman et al., 2011). A 2026 study in JAMA Network Open found that nurse understaffing during the prior 24-hour period was associated with increased odds of in-hospital death, adjusted odds ratio 1.22 (95% CI 1.11–1.34) (Morioka et al., 2026). The Joint Commission identified low nursing staffing as a contributing factor in 24% of 1,609 hospital-reported patient deaths and serious injuries reviewed as of March 2002 (Joint Commission, 2002).
The formula error does not only harm patients. It harms nurses.
Nurses working in hospitals with higher patient loads had significantly elevated odds of burnout and job dissatisfaction (Aiken et al., 2002). In a 2023 survey of more than 7,000 nurses conducted by the American Nurses Foundation, 56 percent reported burnout and 64 percent reported a great deal of job-related stress (American Nurses Foundation, 2023). Chronic understaffing creates conditions in which nurses are systematically unable to meet their professional obligations to patients. Rabin and colleagues identified chronic understaffing as a systemic cause of moral injury in healthcare workers, defined as the persisting distress following exposure to events that transgress deeply held moral beliefs (Rabin et al., 2023). A systematic review of nurses across eight studies found significant positive associations between moral injury and anxiety and depression, and a significant negative association with quality of life (Anastasi et al., 2025). Female nurses die by suicide at nearly twice the rate of the general female population: 17.1 per 100,000 compared to 8.6, with a relative risk of 1.99 (Davis et al., 2021).
None of these outcomes has been traced to the budget methodology, because the methodology has not been questioned. That is the mechanism Vaughan described: the deviation disappears into the background of ordinary practice, and its consequences are attributed to everything except the deviation itself.
At a 20% nonproductive rate, the shortfall is 4 FTEs per 100. At 22%, it is nearly 5. Applied to approximately 1.9 million hospital RN FTEs in the United States (Bureau of Labor Statistics, 2025), the structural gap represents millions of unfilled shifts annually, before a single call-out, turnover event, or census surge is added. Those shifts do not simply go unfilled. They are absorbed through overtime, premium labor, or the suppression of the paid time off and required education the nonproductive rate exists to fund — costs the budget did not account for, because the formula that should have funded them was short from the start.
The formula error is the structural floor beneath every other staffing metric. In 2025, the national RN vacancy rate stood at 8.6%, with the average hospital carrying forty-three unfilled RN positions against its budgeted FTE count (NSI Nursing Solutions, 2026). One in three hospitals reported a vacancy rate of ten percent or higher. On top of that vacancy gap, RN turnover ran at 17.6% for the year. Each departure left the position empty for an average of seventy-eight days before a replacement was hired. New hires enter orientation before reaching independent practice, a period during which preceptor coverage is divided between training and patient care. More than one in five newly hired RNs left within their first year, resetting the cycle before the orientation investment was recovered. These layers of depletion compound on a budget that understated the requirement before any position opened, before any nurse resigned, and before any new hire set foot on the unit.
Units operating at or below their structural baseline for patient care carry no buffer. Every approved vacation day, sick call, and mandatory education requirement, the exact categories the nonproductive rate exists to fund, becomes an unfilled shift that must be absorbed by the remaining staff, covered with premium labor, or left open. The nonproductive rate exists to fund this relief coverage in advance. The correct formula does that. The additive formula does not.
The question is not whether understaffing causes preventable deaths. That is established. The question is how much of the chronic understaffing documented across the nursing literature for decades has been structurally guaranteed by a formula no one corrected.
Seven people died when Challenger broke apart. We do not know how many patients died in units whose budgets were calculated with the wrong formula. That is not because the number is small. It is because the harm has never been concentrated in a single visible event, and events that are not visible are not investigated.
The Challenger parallels are not rhetorical. They are structural.
Vaughan documented that NASA managers had received written warnings from engineers before the launch. The information was available. It was not acted on. The institutional pressure to proceed, to stay on schedule, to avoid disrupting a program with significant political and financial visibility, created conditions in which the engineers who had the math right were overridden by the managers who had the authority.
In nursing finance, the same dynamic has played out across the literature and in professional education settings. The corrected formula, when raised, has been declined not on mathematical grounds but on institutional ones. One workshop instructor told participants that finance would not accept the correct formula, that using it would make them “think you’re trying to cook the numbers,” and that the correct approach simply “is not gonna fly.”
When the same error was raised in a separate CE context, the response was: “I’ve always done it this way and I’ve worked with 3 large consulting firms and they do it this way too.” That is the verbal signature of normalized deviance. The duration of a practice and the number of institutions that share it offered as evidence of its correctness. Vaughan’s framework predicts that response exactly.
The error has also evaded identification in peer review. The additive formula has appeared in refereed nursing journals without challenge, reviewed by editors and peer reviewers trained in the same tradition. When the people responsible for catching errors share the error, peer review cannot function as a corrective mechanism. The formula passes review not because reviewers evaluated it and found it correct, but because they recognized it and assumed it was.
The engineers had the math right. The authority was elsewhere.
The mechanism Vaughan described requires no malice. The Thiokol managers who authorized the launch were not indifferent to the crew. The nurse educators who continue to teach the additive formula are not indifferent to nurses or to patients. The deviation normalizes gradually, through repetition, through institutional endorsement, through the absence of a visible and attributable catastrophe. The formula is used. Budgets are filed. Units appear staffed to finance. The nurse manager knows the unit is short. The charge nurse knows. The patients know. Nurses leave. Patients wait. The shortage is attributed to supply.
The nursing workforce shortage has been debated in terms of both supply and demand for more than eighty years. Reports of an emerging shortage appeared as early as the mid-1930s, driven by rising hospital utilization, more technologically complex patient care, and reductions in nurse working hours (Barbara Bates Center for the Study of the History of Nursing, 2011). Supply responses dominated: expand nursing school admissions, shorten training pipelines, recruit internationally. Demand-side pressure has been recognized too. The complexity of hospitalized patients has grown continuously, hospital admission rates rose substantially across the postwar decades, and the proportion of direct care requiring a registered nurse rather than less trained personnel increased. Nursing positions sit open for weeks and months. Recruitment campaigns, immigration programs, accelerated training pipelines, international hiring, technology solutions: every major structural response has addressed one side or the other. But the most widely used formula for converting care requirements into funded positions has systematically understated the requirement. When facilities correct the calculation, many will find they need dozens more staff than their current vacancy reports reflect. The supply problem is real. The demand has never been accurately counted.
The implications extend beyond staffing budgets. How nursing labor is counted affects how it is reimbursed. That thread belongs in a separate analysis.
The profession that publishes and teaches this formula is the profession that can correct it. That is not a criticism. It is a description of both the problem and the opportunity. No external authority is required. Nursing already applies independent verification to medication dosing calculations because the profession has long recognized that arithmetic errors in clinical practice cause direct, traceable patient harm. A nurse who approximated a weight-based drug calculation rather than performing it correctly would face immediate review. The staffing formula has carried the same class of arithmetic error for sixty-six years without correction. The correction requires only the profession’s willingness to apply to its financial calculations the same standard of mathematical rigor it has always demanded at the bedside.
The check that would catch it takes thirty seconds: multiply the total budgeted FTEs by the productive percentage. If the result equals the productive FTE requirement, the budget is whole. If it does not, it is short before the fiscal year begins.
Every nurse manager, nursing director, and CNO can apply that check to their staffing budgets today.
Challenger is remembered not because the disaster was unforeseeable. It is remembered because it was foreseeable, and it happened anyway. The catastrophe in nursing is also visible, to anyone willing to connect it to its source.
References
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), 1987–1993.
American Nurses Foundation. (2023). Pulse on the nation’s nurses: Mental health and wellness survey (May 2023). American Nurses Association. https://www.nursingworld.org/practice-policy/PulseontheNationsNursesSurvey/nurses-survey-may-2023/
Anastasi, G., Gravante, F., Barbato, P., Bambi, S., Stievano, A., & Latina, R. (2025). Moral injury and mental health outcomes in nurses: A systematic review. Nursing Ethics, 32(3), 698–723. https://doi.org/10.1177/09697330241281376
Barbara Bates Center for the Study of the History of Nursing. (2011). Where did all the nurses go? University of Pennsylvania School of Nursing. https://www.nursing.upenn.edu/nhhc/workforce-issues/where-did-all-the-nurses-go/
Brown, B. (1999, June). How to develop a unit personnel budget. Nursing Management, 30(6), 34–35.
Bureau of Labor Statistics, U.S. Department of Labor. (2025). Occupational employment and wage statistics profiles: 29-1141 Registered nurses (May 2025). U.S. Department of Labor. https://data.bls.gov/oesprofile/?major_group=290000&occupation=291141&measure=01&areas=INDUSTRY,STATE,MSA
Davis, M. A., Cher, B. A. Y., Friese, C. R., & Bynum, J. P. W. (2021). Association of US nurse and physician occupation with risk of suicide. JAMA Psychiatry, 78(6). https://doi.org/10.1001/jamapsychiatry.2021.0154
Graf, C. M. (1984). The operating budget. In S. A. Finkler (Ed.), Budgeting concepts for nurse managers (1st ed., pp. 62–103). W.B. Saunders.
Gunn, M. L. (1960). A method for developing a master staffing plan for the nursing service department. Catholic Hospital Association.
Hunt, P. S. (2018, May). Developing a staffing plan to meet inpatient unit needs. Nursing Management, 49(5), 25–31. https://doi.org/10.1097/01.NUMA.0000532326.62369.9b
Hurst, K. (2003). Selecting and applying methods for estimating the size and mix of nursing teams: A systematic review of the literature commissioned by the Department of Health. Nuffield Institute for Health, Leeds University.
International Council of Nurses. (2006). Safe staffing saves lives [International Nurses Day toolkit]. International Council of Nurses.
Joint Commission on Accreditation of Healthcare Organizations. (2002). Health care at the crossroads: Strategies for addressing the evolving nursing crisis. JCAHO.
Morioka, N., Moriwaki, M., Miyawaki, A., Saville, C., Fushimi, K., & Griffiths, P. (2026). Hospital nurse understaffing and patient mortality, readmission, and length of stay. JAMA Network Open, 9(2), e2558235. https://doi.org/10.1001/jamanetworkopen.2025.58235
Needleman, J., Buerhaus, P., Pankratz, V. S., Leibson, C. L., Stevens, S. R., & Harris, M. (2011). Nurse staffing and inpatient hospital mortality. New England Journal of Medicine, 364(11), 1037–1045. https://doi.org/10.1056/NEJMsa1001025
NSI Nursing Solutions. (2026). 2026 NSI national health care retention & RN staffing report. NSI Nursing Solutions, Inc.
Rabin, S., Kika, N., Lamb, D., Murphy, D., Stevelink, S. A. M., Williamson, V., Wessely, S., & Greenberg, N. (2023). Moral injuries in healthcare workers: What causes them and what to do about them? Journal of Healthcare Leadership, 15, 153–160. https://doi.org/10.2147/JHL.S396659
Strasen, L. (1987). Key business skills for nurse managers. J.B. Lippincott Company.
Thomas, P. L., & Roussel, L. (2020). Procuring and sustaining resources: The budgeting process. In L. A. Roussel, J. L. Harris, & P. L. Thomas (Eds.), Management and leadership for nurse administrators (8th ed., pp. 155–188). Jones & Bartlett Learning.
Vaughan, D. (1996). The Challenger launch decision: Risky technology, culture, and deviance at NASA. University of Chicago Press.
Ward, W. J., Jr. (2016). Health care budgeting and financial management (2nd ed.). Praeger.