The Knowns and Unknowns

Editorial
Nurse Staffing: The Knowns and Unknowns
Jack Needleman, PhD, FAAN
Editorial Board Member
T
HIS ISSUE OF Nursing Economic$ is devoted to examining a diverse set of topics in nurse staffing. These
include examining staffing levels in hospitals and
creating methodologies for staffing at levels that
assure safe and reliable care can be delivered; examining
other aspects of the work environment that can affect performance, including shift lengths; examining the financing
and payment for care; and discussing some of the operational
issues in managing staffing levels. The articles in this issue
(and several in the March/April 2015 issue) also offer insight
into the changing labor market for nursing and look outside
of acute care to reviewing our understanding of the role of
registered nurses (RNs) in assuring patient safety in nursing
homes.
This collection builds on over 2 decades of research and
reflects both the gains in our understanding of nurse staffing
and nurses’ work environment and the tasks still ahead to
effectively manage staffing. In this editorial, I want to review
the key findings from staffing research, identify some outstanding issues in staffing, and place the articles in this special
issue into the context of these open issues. I will discuss nurse
staffing levels and work environment and the delivery of safe
and reliable care, the challenges of modeling safe and appropriate staffing levels, financial analysis of the costs and value
of nursing care, and the changing labor market for nurses.
Nurse Staffing Levels
In 1996, in its report Nurse Staffing in Hospitals and
Nursing Homes: Is it Adequate?, the Institute of Medicine
(IOM) “found that little empirical evidence is available to
support the anecdotal and other informal information that
hospital quality of care is being adversely affected by hospital restructuring and changes in the staffing patterns of nursing personnel….” (p. 9) and that there was a “serious paucity
of recent research on the definitive effects of structural measures, such as specific staffing ratios, on the quality of patient
care…” (p. 9). It set forth a substantial research agenda to fill
the gaps in the understanding of the changing hospital environment and the impact of staffing on patient outcomes
(Wunderlich, Sloan, Davis, & Insitute of Medicine, 1996).
In the years immediately after that report, the levels of
nurse staffing and their association with patient outcomes
became a major focus of health services researchers, with several landmark studies published in leading medical and nursing research journals including the New England Journal of
Medicine (Needleman, Buerhaus, Mattke, Stewart, &
Zelevinsky, 2002), JAMA (Aiken, Clarke, Sloane, Sochalski, &
Silber, 2002), Nursing Research (Blegen, Goode, & Reed,
1998), and Image: The Journal of Nursing Scholarship
(Kovner & Gergen, 1998). The volume of research on staffing
and outcomes increased substantially. For the most part,
these studies have used data on hospital-wide staffing to compare outcomes in high-staffed hospitals to low-staffed hospitals. A 2007 systematic review and meta analysis of the association of nurse staffing levels and patient outcomes identified 96 studies with sufficient rigor to be included in the
analysis (Kane, Shamliyan, Mueller, Duval, & Wilt, 2007).
That review, conducted by researchers at the University of
NURSING ECONOMIC$/January-February 2015/Vol. 33/No. 1
Minnesota, concluded there were statistically significant and
substantively large associations of nurses per patient day and
a wide range of outcomes including mortality, failure to rescue, pulmonary failure, hospital-acquired pneumonia, and
the length of hospital stays, among other outcomes (Kane et
al., 2007). The Kane meta analysis focused on RN staffing, but
studies have also examined total staffing from licensed personnel or hours provided collectively by RNs, licensed practical nurses (LPNs), and nursing assistants. These studies regularly found an association between overall staffing hours
and patient outcomes and the proportion of hours or licensed
hours provided by RNs and outcomes, with more staffing and
greater proportions of RNs associated with better outcomes
(Needleman, Buerhaus, Stewart, Zelevinsky, & Mattke, 2006;
Needleman et al., 2002).
The 1996 IOM report addressed concerns about staffing
in both hospitals and nursing homes. Paralleling the research
on staffing and patient outcomes in hospitals has been substantial research on staffing in nursing homes. Much of this
research has focused on the role of RNs in nursing homes,
generally but not consistently finding an association of
greater RN staffing and better patient outcomes. In the
March/April issue of Nursing Econimic$, Dellefield, Castle,
McGilton, and Spilsbury will note that over the past 8 years,
10 literature or systematic reviews examining these issues
have been published. Dellefield and co-authors (in press)
provide two key services: updating the available systematic
reviews to include literature from 2008-2014, and offering
reflections on how the research on these issues has evolved
in response to earlier criticisms and concerns.
Nurses’ Work Environment
Paralleling the questions about staffing have been questions about the work environment of nurses and its impact
on patient care, safety, nurse satisfaction, and retention.
Indeed, among the earliest focus on work environment was
the original work to identify the factors that made some hospitals better able to retain nurses, that is, magnet hospitals
(McClure, Poulin, Sovie, & Wandelt, 1983; McClure &
Hinshaw, 2002). That work led to the identification of the
factors that contributed to good work environment, subsequently incorporated into the 14 Forces of Magnetism
reflected in the Magnet® program, and into standard measures of work environment such as the Nursing Work Index
(Aiken & Patrician, 2000) and the Practice Environment
Scale (Lake, 2002). It also sparked research demonstrating
that a good nursing work environment is associated with
higher levels of patient safety and fewer adverse events.
Indeed, in a key study, Aiken and colleagues (2011) found a
strong interaction between work environment and staffing,
with the benefits of lower patient-to-nurse ratios sharply
diminished when the work environment was poor.
Work environment, as much as staffing levels, was the
focus of the 2003 IOM report Keeping Patients Safe:
Transforming the Work Environment of Nurses (Page &
Institute of Medicine, 2003). One of the issues examined in
that report was fatigue and the contribution of overtime and
long shift lengths to nurse fatigue. The 12-hour shift has been
5
widely adopted in part because it appears to be preferred by
many nurses but questions have been raised about whether it
is good for either nurses or patients. Moving away from it will
be a challenge, however, and the need and value of moving
from the 12-hour shift needs further assessment. In the
March/April 2015 issue of Nursing Economic$, Martin will
present a pilot study assessing the impact of a shift length
reduction on nurses’ fatigue and between-shift recovery times.
Is the Association of Staffing and Patient Outcomes
Causal?
The general research strategy for assessing the impact of
nurse staffing and work environment on patient outcomes
has been to compare outcomes in hospitals with high staffing
or positive work environments to hospitals with low staffing
or negative work environments. This research is extensive
and utilizes a wide range of measures of staffing and environment, and many different data sets with different patient populations and different hospitals. The findings are robust to
alternative measures and samples. Still, for some observers,
because these studies compare high and low-staffed hospitals, questions linger about whether the results are truly
causal or whether other factors associated with nurse staffing
– physician quality, technology, commitment to high-quality
care, financial resources, differences in patient acuity or need
for nursing – are the real source of the observed association.
Several researchers have attempted to address these concerns. There has been interest in using the natural experiment in California to mandate hospital nurse-to-patient ratios
to assess whether patient care in low-staffed hospitals forced
by regulation to improve staffing has become safer or more
reliable. Some of these studies use the same cross-sectional
high-staffed/low-staffed comparisons of the earlier work,
with the weaknesses associated with this research design.
Some studies exploited the pre-post transition to study
whether staffing has increased at hospitals that were low
staffed before the regulation and whether quality improved in
these hospitals more than quality at hospitals that required
smaller improvements in staffing. One notable study using
this design is by Spetz, Harless, Herrera, and Mark (2013).
The researchers divided California’s hospitals into four quartiles based on their staffing in the pre-mandate period. They
found staffing increased in hospitals and the increases were
higher in hospitals most likely to be subject to the staffing
mandate. Results for changes in patient safety measures were
mixed. Mortality following complications (failure to rescue)
declined more in hospitals with the largest change in staffing
but there was “little evidence of significantly different
changes in other [patient safety indicators] across preregulation staffing levels” (Spetz et al., 2013, p. 393).
While this work is thoughtful and well done, there are
two issues that suggest this study’s negative findings regarding improvement in other patient safety indicators be interpreted guardedly. First, the pre-post comparison implicitly
assumes all else was held equal across hospitals as they
changed staffing. But it is possible other resources than nurses and nursing assistants were reduced or nurses’ workloads
were expanded in other ways as low-staffed hospitals
attempted to accommodate to the demands for more staff.
This would reduce the impact of staffing on outcomes. The
second is the study does not have available critical measures
of nursing work environment. As noted earlier, Aiken and
colleagues (2002) found the association of higher staffing
6
with better outcomes was substantially attenuated in hospitals with poor work environments. If low staffing in the premandate period was also associated with poor work environments, the gains from improved staffing would also be attenuated. The Spetz and co-authors (2013) study is an important
one, trying to make effective use of the California mandate to
study staffing and outcomes, but for the reasons noted, its
conclusions should be treated cautiously. The limitations in
interpreting its findings suggest research studying the
California mandate and staffing regulation in other states will
need to be even more sophisticated to allow us to fully understand how these regulations affect quality of care. This will
include incorporating more data on work environment, nonnurse resources, and hospital financial status in the analysis.
An alternative approach to addressing the causality
question has been to control for other factors by using variations in staff levels within hospitals. In a 2011 study, my colleagues and I used data on nurse staffing and nurse staffing
targets at the unit level in a large academic medical center to
examine whether patients who were exposed to more shifts
on which the targeted RN staffing was 8 hours or more below
target had higher mortality risk (Needleman et al., 2011). We
found mortality risk was increased for those patients, and at
levels that were comparable to those observed in the crosshospital studies based on high and low staffing. Given the
hospital had low baseline mortality, a reputation for high
quality, and that care being studied looked at staffing variations within units, and therefore for care delivered by the
same staff, same physicians, same treatment protocols, and
same technology, this study provides some of the strongest
evidence that the association of staffing and adverse outcomes is causal.
Setting and Achieving Appropriate Staffing Levels
Staffing matters. But even accepting this conclusion, the
question remains: what is the appropriate level of staffing to
assure care is delivered safely and reliably and nurses have
the time to meet the needs of their patients? A wide range of
systems are available for making staffing decisions in hospitals ranging from simple unit-based grids based on census to
data-driven systems that assess individual patient nursing
need using substantial patient-level data and then aggregating these needs up to the unit level. The simple grid systems
are insensitive to day-to-day variations in patient need or the
training and experience of nurses on the unit on a given shift,
while the data-driven systems may be too costly or complex
for smaller hospitals or those with less acutely ill patients.
The need for effective, reliable, credible, usable systems for
projecting appropriate staffing is increasing, driven in part by
state requirements that do not mandate minimum staffing but
do require hospitals to have staff projection systems that have
been vetted by management, staff, and sometimes outside
agencies.
Several authors in this issue address elements of this
challenge. Malloch (2015), Taylor and colleagues (2015), and
Dent (2015) all discuss systems for determining appropriate
staffing levels. Welton and Harper (2015) discuss the challenges of measuring patient nursing acuity, a key component
of assessing necessary staffing. Hill, Higdon, Porter, Rutland,
and Vela (2015) describe some of the management challenges
of balancing staffing and budget. Adams, Kaplow, Dominy,
and Stroud (2015) discuss the management challenges of
assuring targeted staffing levels can be maintained and their
NURSING ECONOMIC$/January-February 2015/Vol. 33/No. 1
success in implementing an internal nursing agency to
accomplish this goal.
The Financial Models to Support Improved Nurse
Staffing
The importance of the business case for staffing is underscored in this issue with Hill and associates (2015) discussing
staffing in the context of budget and the Welton and Harper
(2015) describing value-based payment models for nursing
care. Earlier research examining the business case for nursing,
or assessing whether improved staffing paid for itself in lower
costs due to reduced length of stay and avoided adverse outcomes, priced staffing increases and cost offsets separately
and generally drew two conclusions. The first was that while
a higher RN-to-LPN ratio would pay for itself, increasing nursing hours would not, although savings covered much of the
cost. The second was that the proportion of savings hospitals
would realize depend on the systems for paying hospitals,
with diagnosis-related group or per admission or other bundled payments allowing hospitals to retain savings, while per
diem or proportion of charge-based payment systems shifted
gains to payers. These researchers called for ongoing analysis
of the actual savings hospitals would realize and how these
would be changed by new initiatives in value-based purchasing (Dall, Chen, Seifert, Maddox, & Hogan, 2009; Needleman,
2008; Needleman et al., 2006).
Martsolf and co-authors (2014) examined the net costs of
higher staffing using an alternative model and were more
optimistic about the ability of higher staffing levels to pay for
themselves. They used state hospital discharge data and
state-level hospital staffing reports to examine the impact of
higher staffing levels on adverse events, length of stay, and
patient care costs. As in prior research, they found higher
staffing associated with lower adverse event rates and shorter
length of stay, and that a richer RN-to-LPN mix would be cost
saving. In a notable departure from the prior business case
studies, however, they found higher nursing hours were not
associated with increased patient care costs. If this finding is
confirmed in other studies, it is an important addition to the
debate about whether increases in staffing to assure safe and
reliable care are affordable.
Changing Labor Market for Nursing
Much of the staffing literature focuses on the impact of
staffing on patient care and individual hospitals and nursing
hospitals. But there is other literature that looks at the labor
market for nurses. This research was fueled, in part, by projections of a substantial nursing shortage as the baby boomers
age and need more medical care, as boomer nurses retire, and
as entry into nursing appeared to slow. Auerbach, Buerhaus,
and Staiger have been major contributors to this literature. As
the threat of a nursing shortage has eased due to later retirements and increased class sizes in nursing schools, and as
other labor market issues have emerged such as the growing
demand by hospitals for baccalaureate-prepared nurses, the
focus of their research on the nursing labor market has
expanded. In this issue, Auerbach and colleagues (2015)
examine the changing market for associate degree RNs, finding shifts in demand by hospitals and apparent shifts in
employment for associate degree nurses in long-term care settings.
Enjoy the issue. $
NURSING ECONOMIC$/January-February 2015/Vol. 33/No. 1
REFERENCES
Adams, J., Kaplow, R., Dominy, J., & Stroud, B. (2015). Beyond a Band-Aid® approach:
An internal agency solution to nurse staffing. Nursing Economic$, 33(1), 51-58.
Aiken, L.H., Cimiotti, J.P., Sloane, D.M., Smith, H.L., Flynn, L., & Neff, D.F. (2011).
Effects of nurse staffing and nurse education on patient deaths in hospitals with
different nurse work environments. Medical Care, 49(12), 1047-1053. doi:
10.1097/MLR.0b013e3182330b6e
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.
Aiken, L.H., & Patrician, P.A. (2000). Measuring organizational traits of hospitals: The
Revised Nursing Work Index. Nursing Research, 49(3), 146-153.
Auerbach, D.I., Buerhaus, P.I., & Staiger, D.O. (2015). Do associate degree registered
nurses fare differently in the nurse labor market compared to baccalaureate-prepared RNs? Nursing Economic$, 33(1), 7-12.
Blegen, M.A., Goode, C.J., & Reed, L. (1998). Nurse staffing and patient outcomes.
Nursing Research, 47(1), 43-50.
Dall, T.M., Chen, Y.J., Seifert, R.F., Maddox, P.J., & Hogan, P.F. (2009). The economic
value of professional nursing. Medical Care, 47(1), 97-104.
doi:10.1097/MLR.0b013e3181844da8
Dellefield, M.E., Castle, N.G., McGilton, K., & Spilsbury, K. (in press). The relationship
between registered nurses and nursing home quality: An integrative review
(2008-2014). Nursing Economic$.
Dent, B. (2015). Nine principles for improved nurse staffing. Nursing Economic$,
33(1), 41-44, 66.
Hill, K.S., Higdon, K., Porter, B.W., Rutland, M.D., & Vela, D.K. (2015). Preserving
staffing resources as a system: Nurses leading operations and efficiency initiatives. Nursing Economic$, 33(1), 26-35.
Kane, R.L., Shamliyan, T.A., Mueller, C., Duval, S., & Wilt, T.J. (2007). The association
of registered nurse staffing levels and patient outcomes: Systematic review and
meta-analysis. Medical Care, 45(12), 1195-1204. doi:10.1097/MLR.0b013
e3181468ca3
Kovner, C., & Gergen, P.J. (1998). Nurse staffing levels and adverse events following
surgery in U.S. hospitals. Image: The Journal of Nursing Scholarship, 30(4),
315-321.
Lake, E.T. (2002). Development of the practice environment scale of the Nursing Work
Index. Research in Nursing & Health, 25(3), 176-188.
Malloch, K. (2015). Measurement of nursing’s complex health care work: Evolution of
the science for determining the required staffing for safe and effective patient
care. Nursing Economic$, 33(1), 20-25.
Martin, D.M. (in press). Nurse fatigue and shift length: A pilot study. Nursing
Economic$.
Martsolf, G.R., Auerbach, D., Benevent, R., Stocks, C., Jiang, H.J., Pearson, M.L., ...
Gibson, T.B. (2014). Examining the value of inpatient nurse staffing: An assessment of quality and patient care costs. Medical Care, 52(11), 982-988. doi:
10.1097/mlr.0000000000000248
McClure, M., Poulin, M., Sovie, M., & Wandelt, M. (1983). Magnet hospitals:
Attraction and retention of professional nurses. Kansas City, MO: American
Academy of Nursing.
McClure, M.L., & Hinshaw, A.S. (Eds.). (2002). Magnet hospitals revisited: Attraction
and retention of professional nurses. Washington, DC: American Nurses
Publishing.
Needleman, J. (2008). Is what’s good for the patient good for the hospital? Aligning
incentives and the business case for nursing. Policy, Politics & Nursing Practice,
9(2), 80-87. doi: 10.1177/1527154408320047
Needleman, J., Buerhaus, P., Pankratz, V.S., Leibson, C.L., Stevens, S.R., & Harris, M.
(2011). Nurse staffing and inpatient hospital mortality. The New England
Journal of Medicine, 364(11), 1037-1045. doi:10.1056/NEJMsa1001025
Needleman, J., Buerhaus, P.I., Stewart, M., Zelevinsky, K., & Mattke, S. (2006). Nurse
staffing in hospitals: Is there a business case for quality? Health Affairs
(Millwood), 25(1), 204-211. doi:10.1377/hlthaff.25.1.204
Needleman, J., Buerhaus, P., Mattke, S., Stewart, M., & Zelevinsky, K. (2002). Nurse
staffing levels and the quality of care in hospitals. New England Journal of
Medicine, 346(22), 1715-1722.
Page, A., & Institute of Medicine. (2003). Keeping patients safe: Transforming the work
environment of nurses. Washington, DC: National Academies Press.
Spetz, J., Harless, D.W., Herrera, C.N., & Mark, B.A. (2013). Using minimum nurse
staffing regulations to measure the relationship between nursing and hospital
quality of care. Medical Care Research and Review, 70(4), 380-399. doi:
10.1177/1077558713475715
Taylor, B., Yankey, N., Robinson, C., Annis, A., Haddock, K., Alt-White, A., ... Sales,
A. (2015). Evaluating the Veterans Health Admininstration’s staffing methodology model: A reliable approach. Nursing Economic$, 33(1), 36-40, 66.
Welton, J.M., & Harper, E.M. (2015). Nursing care value-based financial models.
Nursing Economic$, 33(1), 14-19, 25.
Wunderlich, G.S., Sloan, F.A., Davis, C.K., & Institute of Medicine, Committee on the
Adequacy of Nursing Staff in Hospitals and Nursing Homes. (1996). Nursing
staff in hospitals and nursing homes: Is it adequate? Washington, DC: National
Academies Press.
7