1

Integrative Review of the Literature

Lucy Roberts
Bon Secours Memorial College of Nursing
Nursing Research 4322
Dr. Turner
Date: 11/10/20

“I pledge”
 2

Abstract
Purpose: The purpose of this integrative review is to identify the relationship between nurse-to-
patient ratios and in-hospital mortality rates among acute-care patients. Background: A lower
nurse-to-patient ratio is considered a key factor that influences patient mortality rates. The need
for units to adopt adequate registered nurse staffing measures is due to the fact that most of the
in-hospital patient care is primarily completed by RNs. Design and Search Method: The
research design is an integrative review. PubMed and Ovid Discovery Service were the two
computer-based search engines used to locate the articles for the research. A total of four
quantitative articles, along with one qualitative article, were selected and included in this review.
Results and Findings: The evidence gathered for this integrative review came to the
conclusion that a higher nurse-to-patient ratio decreases the rate of mortality of acute-care
patients. Furthermore, the evidence gathered for the integrative review indicates that RN
staffing correlates negatively with mortality after controlling major cofounders in each study.
Limitations: The researcher is inexperienced and does not possess the sufficient knowledge
when performing an integrative review. Implications for practice: higher levels of RN staffing
decreases mortality in acute-care patients. Recommendations for future research: Future
research should focus on developing a scientific method that measures the appropriate RN
staffing level.
 3

Integrative Review of the Literature

The purpose of this integrative review is to identify the relationship between nurse-to-

patient ratios and in-hospital mortality rates among acute-care patients. Currently, there is a

growing amount of research and evidence that supports the notion that more registered nurses

(RNs) available to complete patient care reduces mortality rates of in-hospital patients (West et

al., 2014). The need for a lower nurse-to-patient ratio comes from evidence suggesting patients

are more likely to survive their hospital stay if the nurse-staffing ratios are favorable (Kim et al.,

2020). Therefore, it is important that hospital units begin to adopt policies that require more RN

staffing to decrease the nurse-to-patient load, decrease mortality rates, and improve patient

outcomes (West et al., 2014). The aim of the review is to analyze and discuss published articles

related to the researcher’s PICOT question: In acute care patients, how do units with lower

nurse-to-patient ratio requirements affect mortality rates compared to units with higher nurse-to-

patient requirements?

Design and Methods

The research design is an integrative review. PubMed and Ovid Discovery were the two

computer-based search engines used for this research. The search terms included, ‘nurse,’

‘patient,’ ‘staffing,’ and ‘mortality’. The search engine PubMed yielded 32 articles and Ovid

Discovery yielded 17 articles. The search limits included articles that were published within a

five-year timeframe between the years of 2015 and 2020. The articles had to be published in the

English language and needed to be peer-reviewed qualitative, or quantitative, research studies.

Finally, the articles had to relate to the researcher’s PICOT question: In acute care patients,

how do units with lower nurse-to-patient ratio requirements affect mortality rates compared to

units with higher nurse-to-patient requirements. After setting these parameters, the number of

articles available was significantly reduced, and the date for the publication year was changed

to 2014-2020.
 4

The articles were then chosen based on the following inclusion criteria: acute-care

patients, mortality, and staffing. The research articles were screened and based on the

relevance to the researchers PICOT question and the inclusion criteria. Articles were excluded

from the review if thy failed to meet inclusion criteria. As a result, a total of one qualitative

Dehghan-Nayeri et al (2015) and four quantitative articles Kim et al., 2020; Rochefort et al.,

2020; Wankyo et al., 2018 & West et al., 2014, were selected to be included in this review.

Findings and Results

The results of the five reviewed studies indicate that a higher nurse-patient-ratio

decreases the mortality rate of acute-care patients in the hospital setting (Dehghan-Nayeri et al.,

2015; Kim et al., 2020; Rochefort et al., 2020; Wankyo et al., 2018 & West et al., 2014). A

complete summary of the research articles is presented in Appendix 1. The researcher framed

the review according to the following themes: registered nurse staffing level, surveillance, and

cost.

Registered Nurse Staffing Level

There was a consensus among all five studies that a decrease of in-hospital, RN staffing

levels is associated with increased patient mortality (Dehghan-Nayeri et al., 2015; Kim et al.,

2020; Rochefort et al., 2020; Wankyo et al., 2018 & West et al., 2014). The amount of RN staff

working in the acute-care settings is a detrimental and prominent contributing factor to mortality

in all five articles.

The purpose of Dehghan-Nayeri et al., (2015) was to explore nurses’ experiences of

missed care and the factors affecting it. The researchers used a qualitative methodology with a

content analysis approach. The sample size included 23 nurses working at various units within

Tehran University of Medical Sciences with a minimum work experience of 12 months

(Dehghan-Nayeri et al.,2015). The data was collected through the researchers’ field notes and
 5

recorded in face-to-face interviews, telephonic interviews, and focus group discussions. The

data was collected over a period 12 months during 2012-2013. The data was analyzed using a

conventional content analysis based on Graneheim and Lundman’s (2004) method. The tests

that were used in the study were appropriate for obtaining reliable and valid results (Dehghan-

Nayeri et al., 2015). The study concluded that addressing nursing-staff shortages should be a

priority because is the major determinant in missed-nursing care (Dehghan-Nayeri et al., 2015).

This study is important to the research findings because it shows how nurse-to-patient ratios

can affect patient mortality rate from a nurse’s perspective.

The purpose of the cross-sectional study by Kim et al., (2020) was to explore the

relationships between RN staffing levels and craniotomy inpatient morality. The sample size

included 46,779 patients from 203 hospitals, 43 tertiary hospitals, 153 out of the total 294

general hospitals, and 7 out of the 2868 other smaller hospitals (Kim et al., 2020). The type of

data that was collected from the National Health Insurance (NHI) included information regarding

inpatient mortality after a craniotomy-- from January 1, 2014 to December 31, 2015-- and covers

all of the hospitals (Kim et al., 2020). The data was analyzed using the Chi-square test, the

generalized estimated equation (GEEs), the Gee-based multiple logistic regression, and the

Huber-White sandwich method (Kim et al., 2020). The content analysis was appropriate for

obtaining results. The study concluded that a significant finding from the results demonstrated

that the RN staffing levels had the greatest impact on mortality when compared to the other

morality variables analyzed in the article (Kim et al., 2020).The findings of the study are

important to the research because it demonstrates that a high nurse-to-patient ratios can

improve mortality rates of acute care patients.

The purpose of the longitudinal study by Rochefort et al., (2020) was to add validity to

previous cross-sectional evidence associating RN staffing with patient mortality. The sample

size included 146,349 patients who were followed over 3,478,603 shifts, across 32 distinct
 6

nursing units, for a median follow-up time of 14 shifts (Rochefort et al.,2020). Patient-level data

was collected from the University Health Center (UHC) information systems and linked through

unique patient and hospitalization identifiers (Rochefort et al.,2020). The data was analyzed

using descriptive statistics, the multivariable Cox proportional-hazards-regression model, the

Charleston Comorbidity Index, LAPS, adjusted-hazard ratios (HRs), 95% confidence intervals

(CIs), P-values of <0.005 for 2-tailed Cox, and model-based Wald tests were used for clinical

significance. The content analysis was appropriate for obtaining results. Overall, the clinical

findings of the article suggest that hospital mortality could be reduced by preventing RN

understaffing in different hospital units (Rochefort et al.,2020).

The purpose of the study completed by Wankyo et al. (2018) was to examine the

association between RN nurse staffing and in-hospital mortality of stroke patients. The authors

used a cross-sectional research design, and the data was collected from 2009 the National

Health Insurance (NHI) and Health Insurance Review and Assessment service (HIRA) (Wankyo

et al.). The study had a sample size of 11,819 stroke patients who were treated at 615 hospitals

(Wankyo et al.). Patient-level data collected was based on the stroke patients’ in-hospital,

individual, and clinical variables that have been demonstrated to have the greatest effect on the

patient’s mortality (Wankyo et al.). Data analysis was performed using the following statistical

tests: the independent test, Chi-square test, Logistic Regression and Stata version 14 (Stata

Corp, College Station, Texas), and were appropriate for the study (Wankyo et al.). The tests that

were used in the study were appropriate for obtaining the results. The authors state the findings

of this article are significant because it shows a negative association between nurse staffing and

stroke patients’ mortality even after making adjustments for various variables (Wankyo et al.).

The results of the study are important because it shows that RN staffing has a significant impact

on patient mortality rate in an acute care setting.

 7

West et al., (2014) conducted a cross-sectional, retrospective, risk-adjusted

observational study. The purpose of the study was to investigate if there was a relationship

between the number of staff that are available in ICUs and patients’ chances of survival (West

et al., 2014). The data was collected and merged from the following sources: Intensive Care

National Audit & Research Center (ICNARC) case mix program and an Audit Commission

survey of ICUs conducted in 1998 (West et al., 2014). The sample size included 38,168 patients

from 65 different ICUs and was collected from the merged data set (West et al., 2014). Data

analysis was performed using the following test: Multilevel logistic regression, Imer function part

of the Ime4 package in R version 3.0, and standard logistic regression model (West et al.,

2014). The tests that were used were appropriate for obtaining the results. There were

statistically significant findings among the number of nurses and patient’s risk of mortality,

suggesting that nursing staff availability has the greatest impact on those who are the greatest

risk for mortality (West et al., 2014).

Surveillance

Authors of three of the five reviewed articles used the term ‘surveillance’ and defined it

as the continuous observation of an acute care patient by an RN (Kim et al., 2020; Wankyo et

al., 2018 & West et al., 2014). It is defined as the major factor that connects RN staffing levels to

patient mortality (Kim et al., 2020; Wankyo et al., 2018 & West et al., 2014). Overall, the authors

of the articles concluded that because the RN spends the most amount of time with the patient,

they are more likely to catch detrimental early warning signs and implement life-saving

interventions faster (Kim et al., 2020; Wankyo et al., 2018 & West et al., 2014). Since adequate

RN staffing is key to continuous patient observation, this makes the level of RN staffing a

fundamental factor influencing patient safety (Kim et al., 2020)

A difference noted between the articles, is the West et al., (2014) data source comes

from ICNARC and The Audit Commission. The other two articles data sources are obtained
 8

from NHI and HIRA (Kim et al., 2020; Wankyo et al., 2018). This difference is important because

it may influence the findings, create bias, and limit the generalizability of the findings.

Cost

There was a consensus among four of the five reviewed studies that reducing RN

nursing staff to minimize cost will lead to missed-nursing care and will result in an increased rate

of patient mortality (Dehghan-Nayeri et al., 2015; Kim et al., 2020; Wankyo et al., 2018 & West

et al., 2014). Since RNs make up the majority of the staff population, hospital administrators will

often reduce spending by decreasing the RN hospital staff (Kim et al., 2020 & Wankyo et al.

2018). Evidence has shown that an increase in RN staffing can reduce medical costs because

more RNs on the unit is associated with less missed-nursing care which translates to better

patient outcomes (Kim et al., 2020 & Wankyo et al. 2018). This is significant because it showed

a negative association in the cost and the number of RNs working on the unit.

Discussion and Implications

The articles selected for this integrative review offer an understanding of how nurse

staffing relates to patient mortality. Clarification was obtained from the research findings in

relation to the PICOT question, “In acute care patients, how do units with lower nurse-to-patient

ratio requirements affect mortality rates compared to units with higher nurse-to-patient

requirements?”. The authors of the five research articles emphasize the importance of the

findings in relation to the effect nursing staff levels have on the mortality rates of acute-care

patients. The evidence presented in the quantitative articles is significant because after the

confounding factors in each study were controlled, the results demonstrated a negative

association found among nurse staffing (Wankyo et al., 2018). The implication of these findings

suggests that higher levels of RN staffing throughout the hospital is associated with a decrease
 9

mortality rate in acute-care patients (Wankyo et al., 2018). The research evidence also suggests

that preventing understaffing on in-patient hospital units could drop the patient mortality rate

over the hospital system ubiquitously (Rochefort et al., 2020). Despite the copious amount of

research that has been conducted on this topic, there is little consensus on what an appropriate

method would be to calculate an acceptable RN staffing level (Kim et al., 2020). Future research

should focus on developing a scientific method that measures and quantifies the appropriate RN

staffing level for each unit.

Limitations

In forming this integrative review, the researcher came across several limitations. First,

the researcher is inexperienced and does not possess the requisite knowledge to construct a

credible integrative review. Additionally, a full review was not completed as only five articles

were used for this class assignment. Also noteworthy, none of the five articles selected for this

review were randomized-controlled trials (RCT), and a lack of information regarding the cause-

and-effect of the relationship was also apparent (Fineout-Overholt et al., 2019, p.18). Finally,

due to the insufficient amount of research completed on the researcher’s PICOT question, the

time frame of the articles had to be extended to be within the last six years.

This researcher noted that all five selected articles for this review are considered either

descriptive or qualitative (Level VI evidence); this inherent flaw confirms these studies have a

weak design (Fineout-Overholt et al., 2019, p.18). Furthermore, causation cannot be determined

from the studies with the possibility of confounding variables remaining. All five studies that

were chosen had acquired data that was not obtained through a double-blinded study, and as a

result, there is an increased risk of bias (Fineout-Overholt et al., 2019, p.18). Lastly, all articles

in the review failed to mention the validity and the reliability of measurement tools provided.

Conclusion
 10

The evidence compiled for this integrative review analyzes different hospital units, nurse-

staffing requirements and the effect on patient mortality. Findings discussed throughout the five

articles support that higher-nurse-to-patient ratios equate to lower in-patient mortality rates.

Furthermore, the results of the four quantitative studies demonstrated that when confounding

variables are controlled, there is still a strong negative association between nurse staffing and

patient mortality rates (Wankyo et al., 2018). The findings of the studies support higher nurse-to-

patient ratios throughout all the different hospital units, and in return, has a positive impact on

patient safety and improves health outcomes. In relation to the researchers’ PICOT question,

“In acute care patients, how do units with lower nurse-to-patient ratio requirements affect

mortality rates compared to units with higher nurse-to-patient ratios?”, the evidence suggests

that a higher nurse-to-patient ratio is associated with a decrease in mortality rates in acute-care

patients.
 11

References
Dehghan-Nayeri, N., Ghaffari, F., Shali, M, (2015). Exploring Iranian nurses’ experiences of

missed nursing care: a qualitative study: A threat to patient and nurses’ health. Medical

Journal of the Islamic Republic of Iran, 29, 276. https://mjiri.iums.ac.ir

Kim, Y., Young Kim, S., & Lee, K. (2020). Association between registered nurse staffing levels

and in-hospital mortality in craniotomy patients using Korean national health insurance

data. BMC Nursing, 19 (36) https:// doi.org/10.1186/s12912-020-00430-0

Melnyk, B.M., & Fineout-Overholt, E. (2019). Evidence-based practice in nursing & health care:

A best guide to practice (4th ed.) 18. Philadelphia, PA: Wolters Kluwer.

Rochefort, C., Beauchamp, M., Audet, L., Abrahamowicz, M., & Bourgault, P, (2020).

Associations of 4 nurse staffing practices with hospital mortality. Medical Care, 58, 912-

918. https: doi://0025-7079/20/5810-0912

Wankyo, C., & Sohn, M. (2018). The impact of nursing staffing on in-hospital mortality of stroke

patients in Korea. Cardiovascular Nursing, 33(1), 47-54. https:// doi:

10.1097/JNC.0000000000000415.

West, E., Barron, D., Harrison, D., Rafferty, A., Rowan, K., Sanderson, C. (2014). Nurse

staffing, medical staffing and mortality in intensive care: An observational study.

International Journal of Nursing Studies, 51, 781-794.

https://dx.doi.org/10.1016/j.ijnurstu.2014.02.007
 12

Appendix 1: Summary of the Literature Table

APA Citation Author APA citation: Dehghan-Nayeri, N., Ghaffari, F., Shali, M, (2015). Exploring Iranian nurses’
Qualifications experiences of missed nursing care: a qualitative study: a threat to patient and nurses’ health.
Medical Journal of the Islamic Republic of Iran, 29, 276. https:// doi: https://mjiri.iums.ac.ir
 Dehghan-Nayeri (2015) - PhD, Professor, Nursing and Midwifery Care Research Center,
School of Nursing and Midwifery, Tehran University of Medical Sciences, Tehran, Iran
 Ghaffari (2015) – MSc, RN, PhD, Assistant Professor in Nursing, Ramsar Nursing Care
Research Center, Babol University of Medical Sciences, Mazandaran, Iran
 Shali (2015) – BS, MSCN, Senior Lecturer, School of Nursing and Midwifery, Zanjan
University of Medical sciences, Zanjan, Iran
Background/Problem  The purpose of the study was to explore nurses’ experiences of missed care and the
Statement factors affecting it.
 There are very few previous studies that have explored nurses’ experiences of missed
nursing care.
Conceptual/theoretical  No conceptual or theoretical framework identified
Framework  If a theoretical or conceptual were provided, then it would have helped the authors with
the organization of the study.
Design/  Qualitative methodology with a content analysis approach.
Method/Philosophical  Qualitative content analysis is appropriate for obtaining reliable and valid data results from
Underpinnings textual data, producing new insights, depicting factual information and providing practical
guidelines.
Sample/ Setting/Ethical  The sample included 23 nurses working in various units affiliated with Tehran University
Considerations with a minimum 12-month work experience.
 Purposive sampling was completed.
 Variability was maximized by age, gender, work history, type of employment, type of ward,
and working shift).
 Sampling continued unit data saturation.
 The research protocol was approved by institutional
Major Variables Studied (and  Missed nursing care is an error of omission
their definition), if  Themes: contextual conditions, lack of staff effectiveness, coercion
appropriate  Subthemes: system management failure, team disintegration, lack of motivation,
affectability, silent care, escape, and deprivation.
Measurement Tool/Data  Data were collected through the researchers’ field notes and recorded face-to-face
Collection Method interviews, telephonic interviews and focus group discussions
 13

 Data was collected over 12 months (2012-2013)

Data Analysis  Conventional content analysis based on Graneheim and Lundmans’s (2004) method.
 The contents of each interview were recorded directly after and read several times for
comprehension by the researcher and to ensure correspondence with the study’s
objectives.
 Initial codes were made extracted combined and classified according to comparations.
 Efforts were taken to ensure maximum homogeneity within classes and maximum
heterogenicity between them.
Findings/Discussion  Content analysis results demonstrated contextual conditions, coercion, and deprivation
were conditions affecting care behavior and missed care by staff nurses.
 Several subthemes could be attributed to the three main themes.
 Nursing shortages was the leading cause of missed care.
 Missed nursing care puts patients’ lives at risk.
Appraisal/Worth to practice  Provides lived experiences of nurses’ and their opinion of what in-hospital factors make
them miss care.
 Missed care is related to negative patient outcomes and decrease in patient safety.
 This review focuses on patient mortality rate and nurse staffing.
 14

APA Citation Author APA Citation: Kim, Y., Young Kim, S., & Lee, K. (2020). Association between registered nurse
Qualifications staffing levels and in-hospital mortality in craniotomy patients using Korean national health
insurance data. BMC Nursing, 19 (36) https:// doi.org/10.1186/s12912-020-00430-0
 Kim (2020) – College of Nursing, Eulji University, 553, Sanseong-daero, Sujeong-gu,
Seongnam-si, Gyeonggi-do, Republic of Korea
 Young Kim (2020) – Department of Nursing, Changwon National University, 20
Changwondaehak-ro, Uichang-gu, Changwon-si, Gyeongsangnam-do, Republic of Korea.
 Lee (2020) – Medical Research Collaborating Center, Seoul National University Hospital,
101 Daehak-ro, Jongno-gu Seoul, Republic of Korea
Background/Problem  Background – the level of RN staffing is a fundamental factor that influences patient
Statement safety. Craniotomy patients need intensive care after surgery, and most of the care is
provided by RNs. Adequate RN staffing in essential for providing continuous patient care
with effective surveillance because the RN’s responsibility is to monitor and manage
patient needs. This is the foundation to patient safety.
 The purpose of the study was to investigate the relationships between RN staffing levels
(throughout the entire hospital) and craniotomy inpatient mortality.
 To overcome limitations of previous studies, the authors analyzed RN staffing levels in
general wards, ICUs, and hospitals overall using nationwide data to clarify the relationship
between the level of RN staffing and in-hospital mortality among craniotomy patients.

Conceptual/theoretical  Conceptual framework was proposed by Kane, Shamliyan, Muller, Duval, & Wilt, as well
Framework as variable that Foster developed and the UK Department of Health used for calculating
the hospital-standardized mortality ratio.
Design/  Qualitative study – cross-sectional, descriptive non-experimental design
Method/Philosophical  A literature review was completed, and it looked at previous studies that examined RN
Underpinnings staffing and inpatient mortality rate.

Sample/ Setting/Ethical  Sample size included 46,779 subjects from 203 hospitals.
Considerations  There were 43 out of the total 43 tertiary hospitals, 153 out of the total 294 general
hospitals, and 7 out of the total 2868 other smaller hospitals in Korea.
 15

 Analytical variables were provided by National Health Insurance Service. The variables
were selected based on the exclusion of the following: patients with missing data, patients
transferred to another hospital after craniotomy, patients under 19 or over 86 years of age,
patients treated at hospitals with less than 10 craniotomy cases per year, and patients
treated at hospitals without an ICU.
 The study does not collect and record personal identification information using information
disclosed to the public since review exemption was approved by Eulju University of
human research ethics committees
 The use of fid data for this study was approved by the National Health Insurance Service
Major Variables Studied (and  Variables classified as two levels: patient related and hospital related.
their definition), if  Hospital related – type, ownership, location, number of craniotomy patients, number of
appropriate beds per physician, the RN staffing on general wards, the RN staffing level in the ICU, and
the overall hospital RN staffing level.
 RN staffing level is classified using grades based on bed-to-RN ratio and grade in the
study.
Measurement Tool/Data  The data collection method included a large dataset provided by the National Health
Collection Method Insurance Service (NHIS) that contained hospital information and data on inpatient
mortality after craniotomy from January 1, 2014 to December 31, 2015.
 The national dataset covered all hospitals encompassed National Health Insurance
members and Medical Aid recipients in Korea.
Data Analysis  Data analysis was completed using the chi-square (X^2)
 Patient outcomes, classified as either inpatient death or survival, were analyzed using
GEE-based multiple logistic regression.
 The robust Huber-White sandwich method, was applied because there were more than 40
clusters in GEE regression.
 Significance level P < 0.001 was set
 There was no discussion related to the validity or the reliability of the instrument used to
collect data.

Findings/Discussion  Bed-to-RN ratios of less than 2.5:1 in general wards had the lowest mortality rate.
 The mortality rate of hospitals with more than 4.5 beds per RN exhibited a mortality rate
4.9 times higher than the rate in hospitals with less than 2.5 beds per RN; mortality rates
were higher for hospitals with an overall bed-to-RN ratio
 Regression equation used to control hospital and patient-related factors were entered into
the equation with the level of RN staffing to determine the net effects of the RN staffing
 16

level on in-hospital mortality.

 3 models (1. RN staffing level general ward, 2. RN staffing level in ICU, 3. Overall staffing
level of the hospital)
 The RN staffing level had a consistent effect on in-hospital mortality in all models.

Appraisal/Worth to practice  This review focuses on patient mortality rate and nurse staffing.
 First study to analyze the effects of RN staffing levels on the in-hospital mortality rate of
craniotomy patients using a large national dataset covered all hospitals in Korea.
 17

APA Citation Author APA citation: Rochefort, C., Beauchamp, M., Audet, L., Abrahamowicz, M., & Bourgault, P,
Qualifications (2020). Associations of 4 nurse staffing practices with hospital mortality. Medical Care, 58, 912-
918. https: doi://0025-7079/20/5810-0912
 Rochefort (2020) – School of Nursing, Faculty of Medicine and Health Sciences,
University of Sherbrooke, Research Center Charles-Le Moyne-Saguenay-Lac-Saint-Jean
on Health Innovations and Research Center of the Sherbrooke University Hospital center,
Montreal, QC, Canada
 Beauchamp (2020) - Center for Outcomes Research and Evaluation, Research Institute of
McGill University and Health Center, Montreal, QC, Canada
 Audet (2020) – School of Nursing, Faculty of Medicine and Health Sciences, University of
Sherbrooke, Research Center Charles-Le Moyne-Saguenay-Lac-Saint-Jean on Health
Innovations and Research Center of the Sherbrooke University Hospital center, Montreal,
QC, Canada
 Abrahamowicz (2020) - Center for Outcomes Research and Evaluation, and Department
of Epidemiology, Biostatistics and Occupational Health, Faculty of Medicine, McGill
University, Montreal, QC, Canada
 Bourgault (2020) - School of Nursing, Faculty of Medicine and Health Sciences, University
of Sherbrooke and Research Center of the Sherbrooke University Hospital center,
Montreal, QC, Canada
Background/Problem  The purpose of this longitudinal study was to simultaneously estimate the associations of
Statement 4 nurse staffing practices (RN understaffing, RN education, RN experience, and non-RN
skill) with hospital mortality.
 Previous studies on nurse-staffing practices and patient mortality were based on cross-
sectional studies.
Conceptual/theoretical  No conceptual or theoretical framework identified
Framework If a theoretical or conceptual were provided, then it would have helped the authors with
the organization of the study.
Design/  Qualitative – longitudinal, descriptive non-experimental design.
Method/Philosophical  A review of literature was completed by looking at previous cross-sectional studies that
Underpinnings examined the effect of nurse staffing on patient mortality

Sample/ Setting/Ethical  Sample included a cohort of 146,349 patients that was made up of all adult (18 years and
Considerations older) medical, surgical, and intensive care unit (ICU).
 Patients were admitted between January 1, 2010, and January 15, 2017, to an 800-bed
University Health Center (UHC)
 18

 Patients were followed over 3,478,603 shifts, across 32 distinct nursing units, and for a
median follow-up time of 14 shifts.
 The cohort excluded labor/delivery, psychiatric, and palliative-care patients.
 There was no mention in regard to consent, participant’s autonomy, confidentiality, or
protection from harm.
 Study was approved by the research ethics committee at UHC.
Major Variables Studied (and  In-hospital Mortality (dependent factor)
their definition), if  Registered nurse understaffing (independent factor)
appropriate  Patient exposure to understaffed shifts were calculated for each nursing unit and 8-hour
blocks correlating with night, day, and evening shifts. AKA “unit-shifts”.
 The effects of past exposures, while accounting for the variable that longer hospitals stays
have an increased opportunity to be exposed to understaffed shits.
 Nonregistered Nurse Skill Mix (intendent factor)
 RN education level (independent factor)
 Patient characteristics/ comorbidities (independent factors)
 Nursing unit characteristics (independent factors)
Measurement Tool/Data  Data sources were extracted from UHC’s data warehouse: a relational database that
Collection Method contains demographic, administrative, clinical, and laboratory data obtained from the
UHC/s major information systems and linked using unique patient and hospitalization
identifiers.
 In hospital mortality – the date of in-hospital death was determined using patient
disposition data contained in discharge abstracts.
 Payroll database provided data on nursing staff’s worked hours, RNs’ levels of education,
and experiences. Patient and staffing data were linked by date, nursing unit, and shift.
 RN-to-patient ratios were flagged “understaffed” if the equation (number of patients x 0.25
RN/patient x the number of hours) if the observed hours the RN worked is lower than
expected value by at least 8 hours.
 Time-varying exposure model was used to calculate each patient and shift during his/her
hospitalization, the total proportion of understaffed shifts, among all shifts from hospital
admission to the present shift, avoids immortal time line bias and correctly represented
 Approach measure models admission to the current shift as a time-varying cumulative
average deviation from the corresponding unit-shift mean; this is used to account for
between-unit and between-staff differences in the ratio of non-RN staff on each unit
(higher med/surg).
 Laboratory-based Acute Physiology Score (LAPS) determines severity of illness on
 19

admission (range 0-256); higher score indicates higher risk of death.

 Comorbidities were measured at admission using the Charleston Comorbidity Index,
specially design to quantify the impact of these specific variables on mortality.
 Type of hospital admission and admitting service (medical, surgical, ICU) were collected
from discharge abstracts.
 random effect model – control unit-specific work environment that could prohibit or enable
nursing interventions
 Other time- varying covariates, with fixed effects show: 1) if patient was currently
hospitalized on a medical, surgical, or ICU; 2) current shift of hospitalization (night, day,
evening); 3) whether the current shift was on a weekend or holiday; 4) the number of
patients present at the start of the current unit shift. Cumulative proportion of shifts spent
in an ICU since hospital admission was modeled to account for increased mortality among
ICU patients.
Data Analysis  Data analysis was done using SAS software, version 9.4.
 Descriptive statistics were used to study variables and to assess for any multicollinearity
among the 4 nurse staffing practices.
 Multivariable Cox proportional hazards regression model with a normally distributed
random effect for the current nursing unit of hospitalization
 Time zero – hospital admissions and adjusts the effects of the time-varying cumulative
measures of staffing (RN understaffing, non-RN skill mix, RN education, and RN
experience) for the specific patient and unit.
 Patient censored at discharged or after 90 shifts since hospital (whichever comes first to
limit the analysis of acute in-hospital mortality.
 Squared terms – account for nonlinear effects of age, Charleston Comorbidity Index, and
LAPS. Adjusted Hazard Ratios (HRs) with 95% confidence intervals (CI) and p-value
<0.05 for the 2-tailed Cox model-based Wald tests was used as a criterion for statistical
significance.
 Proportional hazards assumption verified with interaction indicators between variables and
indicators of time.
 Sensitivity Analysis – used Akaike Information Criterion
 There was no discussion related to the validity or the reliability of the instrument used to
collect data.

Findings/Discussion  More deaths occurred on a med/surg floor (higher patient-to-nurse ratio) than ICU (lower
patient-to-nurse ratio).
 20

 Findings suggest hospital mortality could be reduced by increasing RN staff on nursing

units.

Appraisal/Worth to practice  Hospital morality is reduced when RN understaffing is prevented on in-hospital units.
 The review focuses on unit staffing and patient mortality
 21

APA Citation Author APA Citation: Wankyo, C., & Sohn, M. (2018). The impact of nursing staffing on in-hospital
Qualifications mortality of stroke patients in Korea. Cardiovascular Nursing, 33(1), 47-54. https:// doi:
10.1097/JNC.0000000000000415.
 Wankyo (2018) – Graduate School of Public Health, Associate Professor, Seoul National
University, Seoul, South Korea
 Sohn (2018) – Department of Nursing, Professor, Inha University, Incheon, South Korea.
Background/Problem  Stroke is one of the leading causes of the death in Korea, and well-qualified, adequate
Statement nursing force achieves better patient outcomes.
 The purpose of the study is to examine the association between nurse staffing and in-
hospital mortality among stroke patients in a nationally represented sample.
Conceptual/theoretical  No conceptual or theoretical framework identified
Framework  If a theoretical or conceptual assertion were provided, then it would have helped the
authors with the organization of the study.
Design/  Quantitative study – descriptive non-experimental / cross-sectional retrospective study
Method/Philosophical  A literature review was completed by examining previous studies of the impact the effects
Underpinnings of nursing staffing of morality among hospitalized patients.

Sample/ Setting/Ethical  Sample included 11,819 stroke patients who were hospitalized in acute-care settings at
Considerations 615 hospitals.
 There was no mention of participant’s autonomy or protection from harm
 Ethical permission and approval in compliance with the ethical standards of the
institutional research committee and with the 1964 Helsinki Declaration.
Major Variables Studied (and  Stroke patients’ in-hospital mortality, and their individual, clinical, and hospital
their definition), if characteristics
appropriate  Dependent variable: patient mortality
 Independent variables: individual characteristics (age, gender, and insurance type- NHI,
medical aid, or other insurance), clinical characteristics (ischemic stroke, admission via
Emergency department, recipient of surgery, length of hospitalization, medical costs, and
number of comorbidities), and hospital characteristics (hospital type and size; number of
variable diagnostic machines- MRI, PET, and CT; number of physicians and nurses per
100 beds; and geographical region of hospital)
 Overall nurse-to-bed ratio was considered a reasonable measurement of nurse staffing.
Measurement Tool/Data  The authors collected data on the stroke patients’ in-hospital mortality and their individual,
 22

Collection Method clinical, and hospital characteristics. Reliability and validity were not mentioned.
 Originally 700,000 stroke patients were selected by stratified random sampling based on
age (16 strata) and gender (2 strata).
 The patients were excluded from the original sample based on exclusion criteria which
included no CT or MRI that confirmed diagnosis, 19 years old or younger, and patients
lacking information based on individual or hospital characteristics
 Final number of study subjects 11,819 who were treated at 615 hospitals.
Data Analysis  Descriptive statistics included means with standard deviations and frequencies with
percentages
 The independent t test and the chi- square (X^2 test was used to evaluate differences
between patient (individual and clinical) and hospital characteristics at different nurse
staffing levels.
 Logistic regression with adjustment for related factors was used to examin the association
between nurse staffing and in-hospital mortality.
 Logged value of medical costs was used in all analysis to reduce to heteroskedastic effect
of extremely large medical costs.
 All statistical tests were 2 tailed and had a type 1 error of 5%
 The data was analyzed using Stata version 14 (StataCorp, College Station, Texas)
 There was no discussion related to the validity or the reliability of the instrument used to
collect data.
Findings/Discussion  Higher levels of nurse staffing reduce in-hospital stroke mortality in Korea, where stroke
units are not widely available (results more generalizable).
 The negative association found between nurse staffing and stroke, in-hospital mortality
was significant even after adjusting for a wide range of cofounders, such as comorbidities,
hospital size, number of physicians, and medical costs.
 Compared to other studies, the sample used in this article was composed of a
homogenous group, shorter hospital stays, and contained patients younger than 19-years
old as compared to previous studies.
Appraisal/Worth to practice  This review focused on patient mortality rate and nurse-staffing
 Supports a lower patient-to-nurse ratio
 23

APA Citation Author APA Citation: West, E., Barron, D., Harrison, D., Rafferty, A., Rowan, K., Sanderson, C. (2014).
Qualifications Nurse staffing, medical staffing and mortality in intensive care: An observational study.
International Journal of Nursing Studies, 51, 781-794.
https://dx.doi.org/10.1016/j.ijnurstu.2014.02.007
 West (2014) – Faculty of Education and Health, University of Greenwich, United Kingdom
 Barron (2014) – Said Business School and Jesus College, University of Oxford, United
Kingdom
 Harrison (2014) - Intensive Care National Audit and Research Centre, United Kingdom
 Rafferty (2014) - Kings College, London, United Kingdom
 Rowan (2014) - Intensive Care National Audit and Research Centre, United Kingdom
 Sanderson (2014) – London School of Hygiene and Tropical Medicine, United Kingdom
Background/Problem  There is a growing consensus, supported by previous research, that the number of nurses
Statement available for patient care improves patient outcomes in acute-care settings; less research
has been done for patients in the ICU settings
 The purpose of the study is to investigate if there is a relationship between the amount of
staff (nurses, doctors, and support workers) that are available in the ICU, and the patients’
chance of survival.
Conceptual/theoretical  Conceptual framework includes four hypotheses:
Framework  1. Higher number of nurses on the ICU establishment will be associated with lower rates
of patient mortality
 2. Higher numbers of consultants in an ICU will be associated with lower mortality rates.
 3. Higher numbers of support staff in an ICU will be associated with lower mortality rates.
 4. The higher the workload of the unit, the less likely the individual patients are to survive.
Design/  Quantitative study – observational, descriptive non-experimental design
Method/Philosophical  A literature review was done by looking at previous studies which provide information on
Underpinnings Nurse staffing, medical staffing, and workload to show the effects on patient mortality.

Sample/ Setting/Ethical  Sample included 38,168 patients from 65 different ICUs

Considerations  Information about the patients came from the Intensive Care National Audit & Research
Centre (ICNARC) Case Mix Program. An Audit Commission survey of ICUs conducted in
1998 gave information about staffing levels. The merged dataset had information on the
65 ICUs and 38,168 patients.
 Two national datasets to control variables in the analysis (patient condition, and workload
on the unit)
 The participation was voluntary, and the participating units are representative of the
 24

population in terms of geographical spread, unit size, and hospital teaching status
(university vs. non university)
 There was no mention of participants autonomy, confidentiality, protection, or consent
prior to the start of the study.
 Ethical approval was in compliance with London School of Hygiene and Tropical
Medicine ethics committee.
Major Variables Studied (and  Dependent variables: Whether or not the patient survived their stay in ICU and whether or
their definition), if not the patient survived their stay in the acute hospital.
appropriate  Independent variables: size of the clinical workforce (nurses and doctors) the number of
support staff (administrative, clerical, technical, and scientific), and the patient’s own
condition (age, number of comorbidities, number of diagnoses, and the severity of the
illnesses).
 “Surveillance” is the key mechanism linking staffing levels and mortality
Measurement Tool/Data  Data was gathered by The Audit Commission from 65 ICUs in the UK. The survey was
Collection Method conducted in 1998 and gave information regarding staffing levels.
 The data was merged from 2 sources (Audit commission and Intensive Care National
Audit & Research Center [ICNARC]).
 Both data sets were merged by the authors and gave information on 65 ICUs and 38,168
patients.
 The Audit commission gathered data from 69 ICUs in total. Four units were dropped from
the analysis because the number of beds they reported differed by more than three in the
two data sets which created doubt on the reliability of the information about them.
Data Analysis  Statistical analysis was performed using Multilevel logistic regression.
 All of the analysis was completed using the Imer function that is part of the Ime4 package
in the R version.
 Multilevel regression is used for the analysis because it would be inappropriate to
consider patients treated in the same ICU as being independent observations.
 The standard logistic regression model, with the addition of a separate random effect for
each hospital trust, is considered to be a control for unmeasured factors that vary across
units but are constant over time.
 The correlation between the ICU and hospital mortality rates was 0.90
 Multivariable analysis was also completed
 There was no discussion related to the validity or the reliability of the instrument used to
collect data.
Findings/Discussion  Statistical significance interaction between number of nurses and patient’s risk of
 25

mortality.
 The findings suggest that nursing staff availability has the greatest impact on those at
great risk of death.
 Results of multivariable analysis show patients’ risk of mortality increases (ICNARC
model) when a higher proportion of beds in the unit are occupied and when there are
large number of transfers on to the unit.

Appraisal/Worth to practice  Review focuses on nurse staffing and patient mortality

 When controlling other variable to effect patient mortality, nursing staff availability factor
has the greatest impact on patient mortality.