Sie sind auf Seite 1von 17

See

discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/250614997

The relationship between nursing leadership


and patient outcomes: A systematic review

ARTICLE in JOURNAL OF NURSING MANAGEMENT JULY 2013


Impact Factor: 1.5 DOI: 10.1111/jonm.12116

CITATIONS READS

19 1,171

3 AUTHORS, INCLUDING:

Carol Anne Wong Greta G Cummings


The University of Western Ontario University of Alberta
59 PUBLICATIONS 1,373 CITATIONS 165 PUBLICATIONS 2,786 CITATIONS

SEE PROFILE SEE PROFILE

Available from: Greta G Cummings


Retrieved on: 09 October 2015
Journal of Nursing Management, 2013, 21, 709724

The relationship between nursing leadership and patient


outcomes: a systematic review update

1 2 3
CAROL A. WONG PhD, RN , GRETA G. CUMMINGS PhD, RN, FCAHS and LISA DUCHARME BScN, RN, MN
1
Associate Professor, Ontario Ministry of Health and Long-Term Care Nursing Early Career Research Award
Recipient, Arthur Labatt Family School of Nursing, Faculty of Health Sciences, Health Sciences Addition (HSA),
The University of Western Ontario, London, Ontario, 2Professor, Faculty of Nursing, Edmonton Clinic Health
Academy, University of Alberta, Edmonton, Alberta and 3Practice Consultant, Nursing Professional Scholarly
Practice, London Health Sciences Centre (LHSC), London, Ontario, Canada

Correspondence WONG C.A., CUMMINGS G.G. & DUCHARME L (2013) Journal of Nursing Management
Carol A. Wong 21, 709724.
Arthur Labatt Family School of The relationship between nursing leadership and patient outcomes: a
Nursing systematic review update
Faculty of Health Sciences
Room H27, Health Sciences Aim Our aim was to describe the findings of a systematic review of studies that
Addition (HSA) examine the relationship between nursing leadership practices and patient
The University of Western outcomes.
Ontario
Background As healthcare faces an economic downturn, stressful work
1151 Richmond Street
environments, upcoming retirements of leaders and projected workforce
London
shortages, implementing strategies to ensure effective leadership and optimal
Ontario N6A 5C1
patient outcomes are paramount. However, a gap still exists in what is known
Canada
about the association between nursing leadership and patient outcomes.
E-mail: cwong2@uwo.ca
Methods Published English-only research articles that examined leadership
practices of nurses in formal leadership positions and patient outcomes were
selected from eight online bibliographic databases. Quality assessments, data
extraction and analysis were completed on all included studies.
Results A total of 20 studies satisfied our inclusion criteria and were retained.
Current evidence suggests relationships between positive relational leadership
styles and higher patient satisfaction and lower patient mortality, medication
errors, restraint use and hospital-acquired infections.
Conclusions The findings document evidence of a positive relationship between
relational leadership and a variety of patient outcomes, although future testing of
leadership models that examine the mechanisms of influence on outcomes is
warranted.
Implications for nursing management Efforts by organisations and individuals to
develop transformational and relational leadership reinforces organisational
strategies to improve patient outcomes.
Keywords: nursing leadership, patient outcomes, systematic review

Accepted for publication: 30 April 2013

cifically nurse staffing, to patient outcomes, including


Background
patient mortality and adverse events such as medica-
Considerable evidence has emerged in the past decade tion errors, hospital-acquired infections, pressure
linking nursing work environment characteristics, spe- ulcers and falls (Kovner & Gergen 1998, Aiken et al.

DOI: 10.1111/jonm.12116
2013 John Wiley & Sons Ltd 709
C. A. Wong et al.

2002, 2008, Needleman et al. 2002, Tourangeau et al. setting factors, (2) process which is concerned with
2002, Cho et al. 2003, McGillis Hall et al. 2003, mechanisms for coordinating and facilitating patient
Estabrooks et al. 2005, Blegen et al. 2011). However, care and (3) outcomes of care. Originally described as
there is less research examining other work environ- a framework for conceptualising the dimensions of
ment factors, specifically, nursing leadership and its health care practice, Donabedian (1966) postulated
effects on patient outcomes. that each of these core dimensions was a necessary
Effective nurse leaders ensure that appropriate staff- condition for the one that followed. Structures influ-
ing and other resources are in place to achieve safe ence processes and processes influence outcomes. The
care and optimal patient outcomes. At the organisa- SPO framework has been used extensively in examin-
tional level senior nurse executives contribute to stra- ing relationships among organisational structural fea-
tegic directions through their participation in senior tures such as nurse staffing (Cho 2001, McGillis-Hall
level decision-making and their ability to influence & Doran 2007) or leadership (MacPhee et al. 2010)
how nursing is practised and valued (Huston 2008, and nurse (MacPhee et al. 2010) and patient outcomes
Wong et al. 2010). At the department and unit levels, (Cho 2001, McGillis-Hall & Doran 2007). In this
frontline leaders engage nurses in decision-making review we describe the SPO framework to examine the
about patient flow and staffing, quality improvement linkages between leadership and patient outcomes.
activities, and continuous learning opportunities to We isolated the leadership style of nurse leaders in
improve overall care delivery (Page 2004, Tregunno organisations as structure. Leadership was defined
et al. 2009, Thompson et al. 2011). The current eco- broadly as the process through which an individual
nomic downturn, health and safety concerns associ- attempts to intentionally influence another individual
ated with stressful work environments, more leaders or a group in order to accomplish a goal (Shortell &
nearing retirement, and projected workforce shortages Kaluzny 2000, p. 109). We further categorised leader-
are testing the abilities of healthcare leaders in sustain- ship as either relational or task-oriented. Cummings
ing, let alone improving, the level of patient care. In et al. (2010a) explained that leadership styles may be
the challenge for organisations to make practices more broadly characterised as approaches that focus on peo-
cost-effective yet improve outcomes, attract and retain ple and relationships (relationally oriented) to achieve
high-performing staff, and be more responsive to common goals or as styles that focus on structures and
patient needs, effective nursing leadership is critical to tasks (task-oriented). Leaders differ in the extent to
advance agendas for change (Page 2004, Lowe 2005, which they pursue a human relations approach and
Fine et al. 2009). show concern and respect for followers, express appre-
The purpose of this study was to describe the find- ciation and support, and are genuinely concerned for
ings of a systematic review of studies in the literature their welfare (Bass & Stogdill 1990). For example,
that examine the relationship between nursing leader- transformational leadership is a relational leadership
ship practices and patient outcomes. In a previous style in which followers have trust and respect for the
review, we searched electronic databases for the leader and are motivated to go above and beyond nor-
20 years between 1985 and 30 April 2005 and found mal work expectations to achieve organisational goals
only seven studies examining the relationship between (Bass 1985, Bass & Avolio 1994).
nursing leadership and patient outcomes (Wong & Leaders also differ in their attention or focus on the
Cummings 2007). In this review, we updated the evi- groups goals and the means to achieve those goals
dence by adding the results of studies published (Bass & Stogdill 1990). Those with a strong task ori-
between 1 May 2005 and 31 July 2012 to the evi- entation define and organise the roles of followers, are
dence reported in the previous paper and reviewed the concerned with goals, procedures and production,
body of research on the relationship between nursing establish well-defined lines of communication, and are
leadership and patient outcomes. likely to keep their distance psychologically from their
followers (Bass & Stogdill 1990). An example of a
task-oriented style is transactional leadership that
Conceptual framework
emphasises the transaction or economic exchange that
A conceptual framework for the study review was takes place among leaders, colleagues and followers to
developed based on Donabedians (1966) structure accomplish work (Bass & Avolio 1994). The transac-
processoutcome (SPO) framework (see Figure 1). tional leaders role is primarily that of recognising
Three conceptual domains comprise this framework: follower needs and monitoring their role fulfilment
(1) structure which is concerned with organisational or (Bass 1985, Bono & Judge 2004).

2013 John Wiley & Sons Ltd


710 Journal of Nursing Management, 2013, 21, 709724
Systematic review update

Figure 1
Framework for systematic review
(Donabedian 1966).

The process concept in our framework was defined examined the association between leadership and
as the leadership processes or mechanisms by which patient outcomes reported as direct observation of
leaders may contribute to patient outcomes. Examples patient outcomes or extracted from administrative
of these processes may include facilitating work condi- databases. Studies of nurse- or staff-reported patient
tions that promote optimum safe patient care, creating outcomes were excluded. There was no restriction on
open communication with staff to support quality care study design and published English-only articles were
standards, or promoting positive relationships with reviewed. We also required that the relationship
staff that promote work engagement (Cummings et al. between leadership and patient outcomes(s) be
2010a). expressed quantitatively and tested statistically.
Outcome considers the measurable results of care
and we focused exclusively on patient outcomes.
Search strategy and data sources
Patient outcomes were defined as outcomes describing
patient mortality, patient safety outcomes such as the We searched the following eight online bibliographic
incidence of adverse events involving patients (e.g. databases for the period 1 May 2005 to 31 July 2012:
falls, nosocomial infections) or complications during ABI Inform Dateline, Academic Search Complete,
hospitalisation, patient perceptions of satisfaction with Cochrane Database of Systematic Reviews (CDSR),
care, and healthcare utilisation such as length of stay MEDLINE, CINAHL, EMBASE, ERIC and PsychI-
(Doran & Pringle 2011). NFO. Table 1 includes a summary of the search strat-
egy. Some of the databases have changed since the
previous review, however, we were careful to include
Methods all the same sources as in the last review. We also
hand searched the journals, Journal of Nursing Man-
Inclusion criteria for studies
agement, Nursing Research, Nursing Leadership, The
Following the same inclusion criteria used in the pre- Leadership Quarterly, Journal of Nursing Administra-
vious review, we included research studies that exam- tion and Journal of Organizational Behavior as well
ined the relationship between nursing leadership in all as the bibliographies of articles identified for inclusion
types of health care settings and one or more patient in the review. As in the last review we searched
outcomes. Leadership or aspects of leadership includ- research websites for relevant research reports.
ing leadership styles, behaviours, competencies, or
practices were measured as self-reported by leaders,
Study identification and quality review
direct observation of leaders or ratings of leader
behaviours made by followers. A leader was defined The primary author and a research assistant screened
as a nurse in a formal leadership role at any level in a the titles and abstracts of the articles identified by the
health care organisation (e.g. first line, middle and/or search strategy. Articles that potentially met our inclu-
senior leadership/management roles) and who super- sion criteria, and where there was insufficient informa-
vises other nurses. Studies of clinical leadership in tion to make a decision regarding inclusion, were
staff nurses, evaluation of leadership development retrieved and assessed for relevance by the primary
programmes or the testing of leadership instruments author and a research assistant. To assess the method-
were excluded. We included only those studies that ological quality of the final set of articles, we used the

2013 John Wiley & Sons Ltd


Journal of Nursing Management, 2013, 21, 709724 711
C. A. Wong et al.

Table 1
Literature search: electronic databases

Database Search terms


May 2005 to July 2012 Nurse AND Leadership AND Patient Outcome
Main Search Terms/Concepts OR Quality of Healthcare AND Research Number of articles

ABI Inform Dateline all(nurse*) AND all(leader*) AND all(patient outcomes) 59


Academic Search Complete (MM leadership OR TI leader*) AND (TI outcome* 2892
OR MM Outcomes (Health Care)
OR MM Nursing outcomes) AND TX nurs*
CINAHL (MM leadership OR TI leader*) AND 4592
(TI outcome* OR MM Outcomes (Health Care)
OR MM Nursing outcomes) AND TX nurs*
Cochrane nurse* AND Administr* OR manag* 643
OR leader* AND patient satisfaction
OR patient safety OR patient outcome*
OR outcome* OR safety OR infection*
OR fall* OR medication error* OR incident report*
EMBASE Leadership (MESH) AND nurs* 3444
tiab) AND Quality of Healthcare (MESH) AND research (tiab)
ERIC leadership AND quality of health care OR outcomes of treatment 214
MEDLINE Leadership (MESH) AND nurs* (tiab) AND 6627
Quality of Healthcare (MESH) AND research (tiab)
PsychINFO Leadership (MESH) AND nurs* (tiab) AND 1902
Quality of Care (MESH) AND research (tiab)
Manual Search 10
Total 20 383
Total Minus Duplicates 15 180
Papers reviewed 121
Final Selection 13

same quality assessment tool which was adapted and


Table 2
used in the initial review (Wong & Cummings 2007) Summary of quality assessment 20 included quantitative papers
and also used in several published systematic reviews (includes: seven studies from previous review)
(Cummings & Estabrooks 2003, Estabrooks et al.
No. of studies
2003, Lee & Cummings 2008). Each study was
reviewed for quality twice by the lead author. The Criteria Yes No
adapted tool was used to assess the research design, Design
sampling, measurement and statistical analysis of each Prospective studies 20 0
study. Thirteen criteria comprise the tool and a total Used probability sampling 5 15
Sample
of 14 possible points can be assigned for all 13 criteria Appropriate/justified sample size 6 14
(Table 2). 12 items were scored as 0 = not met or Sample drawn from more than one site 15 5
1 = met and the item related to outcome measurement Anonymity protected 19 1
Response rate >60% 8 12
was scored as 2. Based on summed point values, stud-
Measurement
ies were then classified as: weak (04), moderate (59) Reliable measure of leadership 20 0
and strong (1014). Valid measure of leadership 19 1
Effects (outcomes) were observed 14 6
rather than self-reported*
Data extraction and analysis Internal consistency 0.70 when 13 7
scale used
The lead author extracted data from all included arti- Theoretical model/framework used 13 7
Statistical analyses
cles that were double-checked by a research assistant Correlations analysed when multiple 18 2
for accuracy. Data were extracted on the study effects studied
author, journal, country, research purpose and ques- Management of outliers addressed 5 15
tions, theoretical framework, design, setting, subjects,
Weak (04), moderate (59), strong (1014), moderate (n = 3),
sampling method, measurement instruments, reliability strong (n = 17).
and validity, analysis, leadership measures, measures *This item scored 2 points. All others scored 1 point.

2013 John Wiley & Sons Ltd


712 Journal of Nursing Management, 2013, 21, 709724
Systematic review update

of effects on patients, significant and non-significant conducted in the United States, the others being con-
results, discussion and recommendations. Appendix S1 ducted in Canada (n = 4) and Norway (n = 1). The
includes data extraction for the 13 studies retained in studies represented a variety of care settings: inpatient
this updated review along with the data from the acute care units of hospitals (n = 12), nursing homes
seven studies identified in the previous review. (n = 4), dialysis facilities (n = 1), emergency units
Using content analysis and the conceptual frame- (n = 1), home healthcare agencies (n = 1) and neonatal
work as a guide, leadership was categorised as rela- intensive care (n = 1). The findings were combined
tionship or task-oriented and patient outcomes were despite the variety of settings. The study samples repre-
sorted into thematic categories based on their com- sented nurses (n = 10), nurses and managers (n = 6), a
mon characteristics and our previously defined out- cross-section of healthcare professionals including
come categories (Doran & Pringle 2011). Second, nurses (n = 2), a combination of nurses, auxiliary
within each thematic category we identified the pat- nurses and unlicensed care staff (n = 1) and directors of
tern of relationships between relational or task nursing (n = 1). Further details on the characteristics of
focused styles of leadership and changes in specific all 20 included studies can be found in Appendix S1.
patient outcomes. For example, we looked at which
leadership styles were predominantly associated with
Methodological quality of included studies
specific outcomes such as patient satisfaction and if
patient satisfaction increased or decreased as a result The methodological quality of the included articles is
of leadership style. We then analysed the reported summarised in Table 2. All articles used a cross-sec-
relationships between the specific leadership styles or tional design and of the 20 included articles,
practices and the outcomes by category and signifi- 17 (85%) were rated as strong and 3 (15%) as moder-
cance (P < 0.05). We also reviewed studies for the ate. Strengths of the studies included the fact that the
processes by which leadership influenced outcomes majority (n = 19, 95%) used reliable and valid mea-
that were either tested or discussed in studies and or- sures, were multi-sited studies (n = 15, 75%), had
ganised into themes based on common characteristics. acceptable sample sizes (n = 17, 85%) and reported
correlations of multiple effects (n = 18, 90%). In all
but one study, leadership was measured by asking
Results followers to rate the leadership style of their formal
leader. Observed leadership behaviours assessed by a
Description of studies
leaders followers contributes to the construct validity
Article selection for this review is summarised in of the measurement of leadership more effectively
Figure 2. Database and hand searches yielded 20 383 than leader self-report (Dunham 2000, Xin & Pelled
titles/abstracts and of these, 121 were identified as 2003), because self-report measures are subject to
potentially relevant after title and abstract review. social desirability response bias (Polit & Beck 2011).
One of the 121 studies was a published journal article Six (30%) studies used only self-report measures for
(McCutcheon et al. 2009) of one of the retained seven patient outcomes, specifically satisfaction with care.
studies in our previous review (Doran et al. 2004) All other patient outcome measures were collected
which we obtained as a research report from a prospectively in the study or extracted from adminis-
research website. We excluded that article plus 107 trative databases. Eighteen (90%) studies reported
others, leaving only 13 studies meeting our selection using multivariate analysis including multiple regres-
criteria after article reviews. sion, logistic regression, hierarchical linear modelling
We added these 13 articles to the seven articles (HLM), or structural equation modelling (SEM). The
retained in our previously reported systematic review unit of analysis for leadership and patient outcomes
(Wong & Cummings 2007) and described the charac- was the unit/organisational level in 17 studies.
teristics of all 20 included studies. These articles repre- Weaknesses of studies identified in the quality
sent 20 original studies with the exception of the assessment related mainly to sample representative-
Capuano et al. (2005) paper which built on the Houser ness, treatment of outliers, explicit inclusion of a con-
(2003) study by using the same conceptual model, col- ceptual or theoretical framework, response rates and
lecting data on the same variables in another setting reporting study reliabilities for measures used. Only
and then adding the data to the Houser (2003) database five (10%) of studies reported using random sampling
and re-tested the model. All 20 studies were published and 12 (60%) had response rates less than 60%. Six
between 1999 and July 2012 and all but five were (60%) addressed outliers in their data, while seven

2013 John Wiley & Sons Ltd


Journal of Nursing Management, 2013, 21, 709724 713
C. A. Wong et al.

Table 3
Summary of study outcomes: relationship between leadership and patient outcomes

Comment relational (=R) or


Patient outcomes Source Significant findings task-oriented leadership (=T)

Patient satisfaction Doran et al. (2004) Increased Transactional leadership style (R)
Larrabee et al. (2004) NS Transformational leadership (R)
McNeese-Smith (1999) Increased Transformational leadership practices (R)
Gardner et al. (2007) NS Manager ability and support (R)
Kroposki & Alexander (2006) Increased Supervisors work collaboratively with staff (R)
Raup (2008) NS Transformational leadership (R)
Havig et al. (2011) Increased Task-oriented leadership (T)
Patient mortality Houser (2003) Decreased Transformational leadership practices through
increased staff expertise and stability (R)
Pollack and Koch (2003) NS Higher leadership (only respiratory therapists
ratings were significant (R) &(T)
Boyle (2004) NS Inverse association with manager ability
and support (R)
Capuano et al. (2005) Decreased Transformational leadership practices
through increased staff expertise (R)
Cummings et al. (2010b) Decreased High resonant leadership (R)
Tourangeau et al. (2007) Increased Manager ability and support (R)
Patient safety:
(a) Adverse events
Behaviour problems Anderson et al. (2003) Decreased Participative leadership (R)
Restraint use Anderson et al. (2003) Decreased Communication openness (R)
Castle & Decker (2011) Decreased Consensus leadership (R)
Complications of Anderson et al. (2003) Decreased Relationship-oriented leadership and less
immobility formalisation (R)
Fractures Anderson et al. (2003) Decreased Relationship-oriented leadership (R)
Medication errors Houser (2003) Decreased Transformational leadership practices
through increased staff expertise and
stability (R)
Boyle (2004) NS Manager ability and support (R)
Capuano et al. (2005) Decreased Transformational leadership practices
through increased staff expertise (R)
Paquet et al. (2013) Decreased Unit manager support through reduced
absenteeism, overtime and nursepatient
ratio (R)
Vogus & Sutcliffe (2007b) Decreased Trust in leadership increased the effect of
safety organising behaviours (R)
Patient falls Houser (2003) Decreased Transformational leadership practices
through greater staff expertise and
stability (R)
Boyle (2004) NS Manager support ability and support
Capuano et al. (2005) Decreased Transformational leadership practices
through greater staff expertise (R)
Taylor et al. (2012) NS Positive perceptions of hospital
management (R)
Catheter use Castle and Decker (2011) Decreased Consensus leadership (R)
Pressure ulcers Boyle (2004) NS Inverse association with manager ability
and support (R)
Castle and Decker (2011) Decreased Consensus leadership (R)
Flynn et al. (2010) NS Inverse association with manager ability
and support (R)
Taylor et al. (2012) NS Positive perceptions of hospital
management (R)
Inadequate pain Castle and Decker (2011) Decreased Consensus leadership (R)
management
Hospital infections Houser (2003) Decreased Transformational leadership practices
(pneumonia and UTI) through greater staff expertise (R)
Boyle (2004) NS Nurse manager ability and support (R)
Capuano et al. (2005) Decreased Transformational leadership practices
through greater staff expertise (R)

2013 John Wiley & Sons Ltd


714 Journal of Nursing Management, 2013, 21, 709724
Systematic review update

Table 3
Continued

Comment relational (=R) or


Patient outcomes Source Significant findings task-oriented leadership (=T)

(b) Complications
Neonatal PIVH/PVL Pollack & Koch (2003) Decreased Higher leadership (combined ratings of
RNs, MDs & RTs) (R) & (T)
Retinopathy of prematurity(ROP) Pollack & Koch (2003) Decreased Only MDs composite leadership scores
(R) & (T)
Pulmonary embolism/ Taylor et al. (2012) NS Positive perceptions of hospital
deep vein thrombosis management (R)
Patient healthcare utilisation
Number of Gardner et al. (2007) NS Manager ability and support (R)
hospitalizations for
dialysis patients
Hospital readmission Hansen et al. (2011) NS Manager support (R)
rates (AMI, HF or
pneumonia)
Length of hospital stay Paquet et al. (2013) Decreased Manager support through reduced
absenteeism, overtime and nursepatient
ratio (R)

NS, not significant.

(35%) failed to include explicit theoretical or concep- Smith 1999, Houser 2003, Capuano et al. 2005). Gol-
tual frameworks to guide the research and seven eman et al.s (2002) resonant leadership (Cummings
(35%) failed to achieve or report acceptable et al. 2010b), the Bonoma-Slevin leadership model
(a > 0.70) alpha reliabilities of the measures used in (Castle & Decker 2011) and Bass and Stogdills
their studies. (1990) conceptualisation of task vs.-relations-oriented
leadership (Havig et al. 2011) were each used in one
of three studies. Thirteen different tools were used to
Leadership
measure leadership and the most frequently used tools
Leadership was measured in these studies as leadership were Bass and Avolios (1995, 2000) Multifactor
styles, behaviours, competencies and practices. The lead- Leadership Questionnaire (MLQ) (n = 3 studies), and
ership concepts or themes across studies were transfor- Kouzes and Posners (1995) Leadership Practices
mational/transactional leadership (McNeese-Smith Inventory (LPI) (n = 3 studies). In two studies mea-
1999, Houser 2003, Doran et al. 2004, Larrabee et al. sures of leadership were developed for the study: one
2004, Capuano et al. 2005, Raup 2008), relational, par- used items of other standardised measures to assess
ticipative or consensus leadership practices (Anderson task-and relationship-oriented leadership (Havig et al.
et al. 2003, Kroposki & Alexander 2006, Castle & 2011) and the other developed items to measure trust
Decker 2011), task- or relationship-oriented leadership in leadership (Vogus & Sutcliffe 2007b).
(Havig et al. 2011), resonant leadership (Cummings Leadership was measured using a component or
et al. 2010b), leader ability and support (Pollack & subscale of multi-dimensional measures of the work
Koch 2003, Boyle 2004, Gardner et al. 2007, Touran- context in seven studies. In fact this trend was evident
geau et al. 2007, Flynn et al. 2010, Hansen et al. 2011, in six of the 13 retained studies in the more recent
Paquet et al. 2013), trust in leadership (Vogus & Sutc- search. The most commonly used (n = 4 studies) mea-
liffe 2007b) and positive perceptions of leaders (Taylor sure was the manager ability and support subscale of
et al. 2012). Aiken and Patricians (2000) Nursing Work Index-
In nine studies (45%), an explicit leadership theory Revised (NWI-R) (Boyle 2004, Gardner et al. 2007,
was used as the basis for the leadership variable. Bass Tourangeau et al. 2007, Flynn et al. 2010) while in
and Avolios (1994) transformational leadership four studies manager support subscales of Singer
theory was used in three studies (Doran et al. 2004, et al.s (2009) Patient Safety Climate in Healthcare
Larrabee et al. 2004, Raup 2008) and Kouzes and Organizations (PSCHO) survey (Hansen et al. 2011),
Posners (1995) transformational leadership practices Sexton et al.s (2000) Safety Attitudes Questionnaire
model was used in another three studies (McNeese- (SAQ) (Taylor et al. 2012), Karasek et al.s (1998)

2013 John Wiley & Sons Ltd


Journal of Nursing Management, 2013, 21, 709724 715
C. A. Wong et al.

decreased the same adverse patient outcomes. Both


authors postulated that strong leaders tend to retain
higher ratios of competent and proficient staff. Paquet
et al. (2013) findings also revealed the indirect effect of
leadership on patient outcomes (decreased medication
errors and patient length of stay) through reduced
absenteeism, overtime and nurse/patient ratios. In one
study relational leadership (trust) was examined as a
moderator variable, which increased the strength of the
negative relationship between staff safety organising
behaviour and the incidence of medication errors (Vo-
gus & Sutcliffe 2007b).
In the other studies, leadership was examined in
direct association with outcomes and the authors dis-
cussed possible mechanisms by which leadership
styles may have affected outcomes. Cummings et al.
(2010b) pointed to the positive effect on staff perfor-
mance outcomes when leaders present a clear leader-
ship style, convey unambiguous expectations and
engage in open and transparent communication prac-
tices. Other studies suggested that positive leadership
behaviours may be associated with outcomes through
facilitation of more effective teamwork or a clear
vision for quality care (McNeese-Smith 1999, Ander-
son et al. 2003, Pollack & Koch 2003, Doran et al.
Figure 2 2004). Relational leadership styles may also facilitate
Selection of articles for review.
higher levels of staff participation in care decision-
making (Boyle 2004, Kroposki & Alexander 2006,
Job Content Questionnaire (Paquet et al. 2013) and Cummings et al. 2010b, Castle & Decker 2011).
Shortell et al.s (1991) Organisational Assessment Finally, the current focus on patient safety in health-
Scale (Pollack & Koch 2003) were used to assess care organisations was evident in three studies that
leadership. identified the effect of work climate factors including
leadership on safety awareness and organising pro-
cesses as possible mediators of patient safety out-
Process
comes (Vogus & Sutcliffe 2007b, Hansen et al.
As outlined in our conceptual framework, leadership 2011, Taylor et al. 2012).
influences patient outcomes indirectly through pro-
cesses such as making changes in the work context or
Relationship between leadership and patient
influencing staff attitudes, behaviour or performance
outcomes
that may facilitate patient care. In three of the reviewed
studies, the mechanisms or processes by which leader- In this review we found 19 patient outcome variables
ship was indirectly related to patient outcomes were which were grouped into five categories using
explicitly tested using structural equation modelling content analysis: relationship between leadership and
(Houser 2003, Capuano et al. 2005, Paquet et al. (1) patient satisfaction, (2) patient mortality, patient
2013). Houser (2003) found that transformational safety outcomes, (3) adverse events and (4) complica-
leadership practices were positively related to staff tions, and (5) patient healthcare utilisation. A sum-
expertise and negatively related to staff turnover, both mary of findings is presented in Table 3. In 30%
of which contributed to reduced patient mortality, hos- (n = 6) of studies patient outcomes, primarily patient
pital acquired infections, medication errors and patient or family satisfaction, were collected prospectively by
falls. Capuano et al. (2005) added to these findings by researchers. In one study (Pollack & Koch 2003)
showing that transformational leadership practices patient mortality and complications were collected
were also associated with staff expertise, which in turn from clinical charts and in all other studies patient

2013 John Wiley & Sons Ltd


716 Journal of Nursing Management, 2013, 21, 709724
Systematic review update

Figure 3
Structure-ProcessOutcome relation-
ships suggested by review findings.

outcomes were collected from administrative databas- Patient mortality


es. Of the individual outcomes, patient mortality In four of six studies leadership was significantly asso-
(n = 6 studies) and medication errors (n = 5 studies) ciated with patient mortality (Houser 2003, Capuano
were the most frequently examined outcomes. Over et al. 2005, Tourangeau et al. 2007, Cummings et al.
all studies, a total of 43 relationships between leader- 2010b). Transformational and resonant leadership
ship and patient outcomes were examined and 63% were associated with lower patient mortality in three
(n = 27) of these were significant. While 26 relation- studies while, contrary to hypothesis, leadership was
ships between leadership and positive patient out- associated with higher mortality in one study (Touran-
comes were significant, one relationship showed the geau et al. 2007). Tourangeau et al. suggested that
opposite of the expected results. managers with larger spans of control may have been
hampered in their ability to provide direct support to
Patient satisfaction nursing staff. However, there was another contradic-
The number of studies (n = 7) relating leadership tory finding in that lower mortality was associated
practices to patient satisfaction was more than dou- with higher nurse burnout.
bled in this review. The results showed significant
associations between leadership and increased patient Patient safety outcomes: adverse events
satisfaction in four studies (McNeese-Smith 1999, A total of nine studies addressed ten types of out-
Doran et al. 2004, Kroposki & Alexander 2006, comes in this category. The strongest relationship was
Havig et al. 2011). In three others (Larrabee et al. between leadership and medication errors, as four of
2004, Gardner et al. 2007, Raup 2008) the results five studies showed significant negative relationships.
were not significant. One of these studies had a small Transformational leadership (Houser 2003, Capuano
sample and had the lowest quality rating (Raup et al. 2005), manager support (Paquet et al. 2013)
2008). Relational leadership was associated with and trust in leadership (Vogus & Sutcliffe 2007b)
patient satisfaction in two studies (McNeese-Smith were all associated with lower medication errors.
1999, Kroposki & Alexander 2006) while Havig et al. Patient falls were examined in four studies and the
(2011) found that family satisfaction with resident results were mixed since two studies (Houser 2003,
care was significantly positively related to task- Capuano et al. 2005) showed significantly decreased
oriented leadership style of nursing home ward man- patient falls related to transformational leadership,
agers. Similarly, Doran et al. (2004) found that the while in two other studies (Boyle 2004, Taylor et al.
transactional leadership style, which can also be cate- 2012) manager support was not significantly related
gorised as task-oriented, was related to increased to fall rates. A lower incidence of pressure ulcers was
patient satisfaction. Doran et al. offered the explana- significantly associated with leadership in only one of
tion that transactional leaders may facilitate patient three studies (Castle & Decker 2011). Two studies
care by providing direction, clarification of tasks and (Anderson et al. 2003, Castle & Decker 2011) found
clear work expectations. significant relationships between positive leadership

2013 John Wiley & Sons Ltd


Journal of Nursing Management, 2013, 21, 709724 717
C. A. Wong et al.

styles (consensus and participative) and lower restraint and patient outcomes by extending the search criteria of
use in nursing homes. In two of three studies examin- the previous review to include studies published
ing hospital-acquired infections (pneumonia and uri- between 1985 and 30 July 2012. The 13 additional
nary tract infections) transformational leadership was studies published after our first systematic review
associated with lower infection rates (Houser 2003, almost triples the number of studies specifically examin-
Capuano et al. 2005). ing the relationship between nursing leadership and a
variety of patient outcomes. The totality of the 20
Patient safety outcomes: complications included studies reflects positive trends in terms of
Three types of complication outcomes were research design and methods. Most studies were multi-
addressed in two studies and thus there were few sited, incorporated multiple levels of analysis, used
studies for each type of complication limiting the more advanced statistical procedures (e.g. HLM, SEM)
ability to draw conclusions. Pollack and Kochs and examined the relationship leadership and patient
(2003) study conducted in neonatal intensive care outcomes in a wider variety of clinical settings, although
settings found a reduced incidence of neonatal peri- the majority were conducted in acute care. What was
ventricular haemorrhage/periventricular leukomalacia disappointing was that less than half of the studies used
(PIVH/PVL) associated with higher leadership rat- explicit leadership models, very few studies examined
ings. Taylor et al. (2012) found no relationship mechanisms of leadership influence on outcomes, there
between leadership and pulmonary embolism/deep was an over-reliance on cross-sectional designs and con-
vein thrombosis. siderable heterogeneity of patient outcomes and clinical
settings precluding greater synthesis of findings.
Patient healthcare utilisation The findings provide support for the assertion that
A new category of patient outcomes was added since relational leadership practices are positively associated
three studies addressed patient healthcare utilisation with some categories of patient outcomes. The Donabe-
indicators, specifically the number of hospitalisations, dian (1966) SPO model provided a useful approach to
hospital readmissions and hospital length of stay as out- organising findings which are summarised in a diagram
comes related to work environment factors including depicting suggested linkages between relational or task
leadership (Gardner et al. 2007, Hansen et al. 2011, oriented leadership, processes of leadership and three of
Paquet et al. 2013). Healthcare utilisation measures the patient outcomes where findings were supported in
reflect services or resources consumed in managing terms of number of studies and direction of effect (Fig-
patients health-related needs (Clarke 2011). Gardner ure 3). The findings highlighted a key relationship
et al. (2007) proposed that patient hospitalisations are between relational leadership and the reduction of
considered important indicators of the general health adverse events, specifically, medication errors, possibly
status of patients who are on dialysis and may be con- through leaders influence on human resource variables
sidered a reasonable nurse-sensitive quality indicator in that may be connected to patient care outcomes, staff
dialysis settings. Hansen et al. (2011) claimed that the expertise, turnover, absenteeism, overtime and nurse to
frequency of hospital readmission rates reflected an patient ratios. There were also promising trends in find-
inadequate patient safety process which means that ings for restraint use and hospital-acquired infections.
hospitals with poorer safety cultures would be expected Findings on mortality outcomes were strong showing a
to exhibit higher levels of hospital readmissions. Man- significant negative relationship between leadership and
ager support was included as one element of patient patient mortality in three of six studies. This important
safety culture. Both studies did not demonstrate signifi- connection may suggest that effective nursing leader-
cant findings for the effects of leadership on these two ship is essential to the creation of practice environ-
healthcare utilisation outcomes. However, Paquet et al. ments, with appropriate staffing levels, resources and
(2013) found that manager support was associated with care processes that support nurses in preventing unnec-
a lower patient length of stay through the human essary deaths (Wong & Cummings 2007). Finally, there
resource indicators of lower absenteeism, overtime and was a significant positive relationship between both
nurse to patient ratio. relational and task-oriented leadership and patient sat-
isfaction. This finding may indicate that some elements
Discussion of each style are needed to ensure care processes that
contribute to satisfied patients such as clear standards
In this systematic review we examined empirical evi- of care and role expectations as well as collaborative
dence on the connections between nursing leadership working relationships.

2013 John Wiley & Sons Ltd


718 Journal of Nursing Management, 2013, 21, 709724
Systematic review update

cation in healthcare. These theories focus on the lea-


Implications for leadership theory
ders ability to attend to the individuals emotions and
Despite progress in acknowledging the inclusion of the outcomes of these behaviours for individuals well-
leadership as an important factor in theoretical and being and performance (Cummings 2004). Likewise,
analytical models of antecedents to patient outcomes, leadermember exchange (LMX) theory has received
there is a pressing need for much stronger conceptuali- little attention in healthcare despite a large empirical
sations of leadership that clearly define leadership base in other disciplines and the potential to increase
practices and specify possible direct and indirect understanding of the linkages between leadership, pro-
mechanisms by which leaders affect individuals and cesses and patient outcomes (Graen & Uhl-Bien
outcomes (Avolio 2007, Gilmartin & DAunno 2007). 1995). In this theory, the nature and quality of the
Although some of the studies included multi-level relationship between the leader and the follower that
analyses of individual-, unit- and organisational-level forms over time is posited to play a vital role in
variables, the articulation of mechanisms by which employee responses to their work environments.
leadership influences the various levels was often inad- Research has linked LMX quality to positive individ-
equate. While an important concept, the definition ual and organisational outcomes, including job perfor-
and measurement of leadership as manager ability or mance (Laschinger et al. 2007).
support is a somewhat narrow notion of leadership, as
opposed to more complex explanations of leadership
Implications for future research
behaviour and influence strategies suggested in other
theories such as transformational or resonant leader- Future studies need to include longitudinal, quasi-
ship theory. experimental or experimental designs that address the
Transformational leadership was the most fre- effects of leadership on patient outcomes. In order to
quently applied leadership theory in the reviewed examine the causal association between leadership and
studies. Attention to other leadership theories that patient outcomes, interventional and/or longitudinal
may have relevance to nursing and healthcare are studies with multiple observations are essential. It is
worthy of further application for the potential mediat- concerning that even though there was a substantial
ing processes they propose between leadership and increase in the number of studies addressing leadership
outcomes (Mark et al. 2004, Gilmartin & DAunno and patient outcomes all of the studies were cross-
2007). For example, authentic leadership (Avolio sectional in design. The trend seen in this review for
et al. 2004) is an emergent leadership approach from studies examining leadership and patient outcomes in
the field of positive organisational behaviour that other than acute care settings, e.g. nursing home, home
highlights the importance of examining the context healthcare and community settings needs to be contin-
and the influence of followers in the leaderfollower ued. The majority of studies were multi-sited which is
dynamic. This relational leadership approach is important for obtaining adequate sample sizes for
grounded in the leaders positive psychological capaci- multi-level analysis. However, these studies also present
ties, honesty and transparency, strong ethics and a significant challenge in terms of collecting reliable
behavioural integrity (Avolio et al. 2004, Wong & and valid data from multiple sites. Likewise, the use of
Cummings 2009). The utility of this theory is that it random sampling procedures would significantly
emphasises possible mechanisms through which lead- strengthen studies. In this review the majority of studies
ership influences performance, and how followers used convenience sampling. One notable issue was that
shape leadership within and between various organisa- only eight (40%) of studies had a response rate of 60%
tional contexts, climates and cultures (Peterson et al. or more, despite convenience sampling. Additional
2012). Testing of various propositions about mecha- activities to increase response rates would improve the
nisms of leader influence on individual, unit and reliability of the results and strengthen data analysis,
organisational outcomes suggested by the theory although the challenge of accessing subjects in increas-
might include the effects of leader positivity on ingly complex, dynamic and fast-paced healthcare set-
followers engagement and performance in new work- tings must be acknowledged.
place initiatives that promote patient safety and better
outcomes.
Implications for practice
Resonant leadership theory (Boyatzis & McKee
2005) based on the concept of emotional intelligence The findings from this review underscore the value of
(Goleman et al. 2002) is also worthy of greater appli- relational leadership styles in the modern healthcare

2013 John Wiley & Sons Ltd


Journal of Nursing Management, 2013, 21, 709724 719
C. A. Wong et al.

workplace and align with findings from other reviews


Limitations
where relational leadership styles were positively and
indirectly related to nurses motivation to perform Even though meticulous methods were employed in
(Brady-Germain & Cummings 2010), improved work this review there were limitations. First, the variety
environments and outcomes for nurses (Cummings of leadership and outcome measures and the hetero-
et al. 2010a) and nurse retention (Cowden et al. geneity of samples and settings precluded meta-analy-
2011). These findings suggest that a complex interplay sis procedures and limited the consolidation of
of associations between the relational practices of for- findings. While an attempt was made to review some
mal nursing leaders to provide vision, support, staffing grey literature (e.g. searching research websites) we
resources (Kane et al. 2007) and leadership, with the did not search all grey literature and did not include
health, competencies, abilities, knowledge, skills and unpublished dissertations, and, as such, this review
motivation of nurses, are integral to the achievement update may not be representative of all the relevant
of better patient outcomes. work in the field. The inclusion of studies published
Thereby, recruiting and retaining individuals into only in the English language may have also excluded
leader roles with the requisite emotional intelligence other potentially informative studies. Studies may
competencies that underpin relational leadership have been excluded that contained leadership data
styles are critical to effective performance at all lev- but because it was not purposely hypothesised or dif-
els of organisations (Cummings 2004, Young-Ritchie ferences in outcomes by leadership style were not
et al. 2009). Relationally oriented leaders contribute examined those results were not accessible. The vari-
to positive practice settings and staff work engage- ability in the conceptualisation and measurement of
ment by providing support and encouragement, posi- leadership across studies may limit the validity and
tive and constructive feedback, open and transparent generalisability of findings. All of the studies included
communication and individual consideration. Creat- were cross-sectional in design, limiting interpretations
ing opportunities for meaningful dialogue between of causality to the evidence of co-variation in the
leaders and clinical nurses is necessary to discuss study variables and foundational theoretical associa-
patient care issues that could impede patient safety. tions. The theoretical underpinnings and causal
While this is challenging in the current high-paced, understandings of studies were not reported by all
and heavy meeting-laden managerial roles, it remains researchers, which may have influenced appropriate
a priority that cannot be overlooked. Nurses must synthesis of findings.
be provided with the opportunity and staffing
resources to monitor patients conditions and address
Conclusion
their education needs regarding self-care, symptom
management and other factors related to patient In this systematic review update we identified an
empowerment. increasing body of research findings relating to the
The connection noted between supportive leadership relationship between nursing leadership and patient
styles and positive patient safety outcomes may point outcomes. However, the prominence of cross-sectional
to the importance of unit leaders understanding of designs and the heterogeneity in patient outcome vari-
patient care processes and the role of nurses and other ables, samples/settings and leadership measures mean
healthcare providers in promoting better outcomes. that robust evidence to support specific leadership
Thompson et al.s (2011) recent study provided evi- styles that predict specific patient outcomes is limited.
dence that when leaders demonstrated higher rela- The current evidence reinforced findings from the pre-
tional leadership, the staff on their units reported vious review with respect to the positive relationships
more positive patient safety climates. The ability of between relational leadership styles and patient
leaders to promote a safe workplace is governed by satisfaction and improved patient safety outcomes.
their knowledge of patient care needs, their level of Specifically, the current evidence suggests a clear rela-
relational skills and their capacity to recognise and tionship between relational leadership styles and lower
implement effective safety practices (Tregunno et al. patient mortality and reduced medication errors,
2009, Thompson et al. 2011). Additionally, the restraint use, and hospital-acquired infections. Future
development of safety cultures through leadership studies of a longitudinal and interventional nature
interventions, which include managers interdisciplin- must be conducted in a variety of settings with more
ary walkabout safety rounds, have been linked to diverse and randomly selected samples. In addition, the
improved outcomes (Ginsburg et al. 2005). development and testing of stronger conceptual models

2013 John Wiley & Sons Ltd


720 Journal of Nursing Management, 2013, 21, 709724
Systematic review update

of leadership and the mechanisms of influence on patient Bass B.M. & Avolio B. (1995) Multifactor Leadership Ques-
outcomes is warranted to advance knowledge of the tionnaire (MLQ): Sampler Set Manual. Mind Garden,
Redwood City, CA.
complex contextual and multivariate influences on the
Bass B. & Avolio B. (2000) Multifactor Leadership Question-
relationship between leadership and patient outcomes. naire. Mind Garden, Inc., Redwood City, CA.
Bass B.M. & Stogdill R.M. (1990) Bass and Stogdills Hand-
book of Leadership: Theory, Research and Applications. Free
Sources of funding Press, New York, NY.
We gratefully acknowledge salary funding support Blegen M.A., Goode C.J., Spetz J., Vaughn T. & Park S.H.
(2011) Nurse staffing effects on patient outcomes: safety-net
through the Ontario Ministry of Health and Long-
and non-safety-net hospitals. Medical Care 49 (4), 406414.
Term Care Nursing Early Career Research Award and Bono J.E. & Judge T.A. (2004) Personality and transforma-
in-kind support from the University of Western tional and transactional leadership: a meta-analysis. Journal
Ontario. of Applied Psychology 89, 901910.
Bonoma T. & Slevin D.P. (1978) Executive Survival Manual.
CBI Publishing Company, Boston, MA.
Ethical approval Boyatzis R. & McKee A. (2005) Resonant Leadership. Harvard
Business School Press, Boston, MA.
Ethical approval was not required for this paper. *Boyle S.M. (2004) Nursing unit characteristics and patient out-
comes. Nursing Economics 22 (3), 111123.
Brady-Germain P. & Cummings G.G. (2010) The influence of
References nursing leadership on nurse performance: a systematic litera-
ture review. Journal of Nursing Management 18, 425439.
*Capuano T., Bokovoy J., Hitchings K. & Houser J. (2005) Use
*Included studies of a validated model to evaluate the impact of the work envi-
Aiken L.H. & Patrician P.A. (2000) Measuring organizational ronment on outcomes at a magnet hospital. Health Care
traits of hospitals: The Revised Nursing Work Index. Nursing Management Review 30 (3), 229236.
Research 49, 146153. *Castle N.G. & Decker F.H. (2011) Top management leader-
Aiken L.H., Sochalski J. & Lake E. (1997) Studying outcomes ship style and quality of care in nursing homes. Gerontologist
of organizational change in health services. Medical Care 35 51 (5), 630642.
(11), NS6NS18. Cho S.H. (2001) Nurse staffing and adverse patient outcomes: a
Aiken L.H., Clarke S.P., Sloane D.M., Sochalski J. & Silber systems approach. Nursing Outlook 49 (2), 7885.
J.H. (2002) Hospital nurse staffing and patient mortality, Cho S.H., Ketefian S., Barkauskas V.H. & Smith D.G. (2003)
nurse burnout, and job dissatisfaction. Journal of the Ameri- The effects of nurse staffing on adverse events, morbidity,
can Medical Association 288 (16), 19871993. mortality, and medical costs. Nursing Research 52 (2),
Aiken L.H., Clarke S.P., Sloane D.M., Lake E.T. & Cheney T. 71Y79.
(2008) Effects of hospital care environment on patient mortal- Clarke S.P. (2011) Healthcare utilization. In Nursing Outcomes:
ity and nurse outcomes. Journal of Nursing Administration The State of the Science, 2nd edn(D.M. Doran ed), pp.
38 (5), 223229. 487508. Jones & Bartlett Learning, Sudbury, MA.
Alexander J.W. & Randolph W.A. (1985) The fit between tech- Cowden T., Cummings G. & Profetto-McGrath J. (2011) Lead-
nology and structure as a predictor of performance on nurs- ership practices and staff nurses intent to stay: a systematic
ing units. Academy of Management Journal 28, 844859. review. Journal of Nursing Management 19, 461477.
Anderson R.A., Ashmos D.P., McDaniel R.R. & Hsieh P.C. Cummings G.G. (2004) Investing relational energy: the hall-
(1997) Complexity theory view of participation in decision mark of resonant leadership. Canadian Journal of Nursing
making: conceptual and empirical analysis. Paper presented at Leadership 17 (4), 7687.
the Proceedings 1997 Academy of Management Annual Meet- Cummings G.G. & Estabrooks C.A. (2003) The effects of hos-
ing, Boston, MA. pital restructuring that included layoffs on individual nurses
*Anderson R.A., Issel L.M. & McDaniel R.R. (2003) Nursing who remained employed: a systematic review. International
homes as complex adaptive systems. Nursing Research 52 (1), Journal of Sociology & Social Policy 8 (9), 853.
1221. Cummings G.G., MacGregor T., Davey M. et al. (2010a) Lead-
Avolio B.J. (2007) Promoting more integrative strategies for lead- ership styles and outcome patterns for the nursing workforce
ership theory-building. American Psychologist 62 (1), 2533. and work environment: a systematic review. International
Avolio B.J., Gardner W.L., Walumbwa F.O., Luthans F. & Journal of Nursing Studies 47, 363385.
May D.R. (2004) Unlocking the mask: a look at the process *Cummings G.G., Midodzi W.K., Wong C.A. & Estabrooks
by which authentic leaders impact follower attitudes and C.A. (2010b) The contribution of hospital nursing leadership
behaviours. Leadership Quarterly 15, 801823. styles to 30-day patient mortality. Nursing Research 59 (5),
Bass B.M. (1985) Leadership and Performance Beyond Expecta- 331339.
tions. Free Press, New York, NY. Donabedian A. (1966) Evaluating the quality of medical care.
Bass B.M. & Avolio B.J. (1994) Improving Organizational Millbank Memorial Fund Quarterly 44 (3), 166206.
Effectiveness Through Transformational Leadership. Sage, Doran D.M. & Pringle D. (2011) Patient outcomes as an
CA. accountability. In Nursing-Sensitive Outcomes(D. Doran ed),

2013 John Wiley & Sons Ltd


Journal of Nursing Management, 2013, 21, 709724 721
C. A. Wong et al.

pp. 127. Jones and Bartlett Publishers International, Missis- Huston C. (2008) Preparing nurse leaders for 2010. Journal of
sauga, ON. Nursing Management 16, 905911.
*Doran D., McCutcheon A.S., Evans M.G. et al. (2004) Impact Kane R.L., Shamliyan T.A. & Mueller C. (2007) The associa-
of the Managers Span of Control on Leadership and Perfor- tion of registered nurse staffing levels and patient outcomes:
mance. Canadian Health Services Research Foundation, systematic review and meta-analysis. Medical Care 45 (12),
Ottawa, ON. 11951204.
Dunham T.J. (2000) Nurse executive transformational leader- Karasek R., Brisson C., Kawakami N., Houtman I., Bongers P.
ship found in participative organizations. Journal of Nursing & Amick B. (1998) The Job Content Questionnaire (JCQ):
Administration 30 (5), 241250. an instrument for internationally comparative assessments of
Estabrooks C.A., Floyd J.A., Scott-Findlay S., OLeary K.A. & psychosocial job characteristics. Journal of Occupational
Gushta M. (2003) Individual determinants of research utiliza- Health Psychology 3 (4), 322355.
tion: a systematic review. Journal of Advanced Nursing 43 Kirkevold . & Engedal K. (2006) The quality of care in Nor-
(5), 506520. wegian nursing homes. Scandinavian Journal of Caring Sci-
Estabrooks C.A., Midodzi W.K., Cummings G.G., Ricker ences 20 (2), 177183.
K.L. & Giovannetti P. (2005) The impact of hospital nurs- Kouzes J. & Posner B. (1988) The Leadership Challenge: How
ing characteristics on 30-day mortality. Nursing Research to Get Extraordinary Things Done in Organizations. Jossey-
54 (2), 7484. Bass, San Francisco, CA.
Fine B.A., Golden B., Hannam R. & Morra D. (2009) Leading Kouzes J.M. & Posner B.Z. (1995) The Leadership Challenge:
lean: a Canadian healthcare leaders guide. Healthcare Quar- How to Get Extraordinary Things Done in Organizations,
terly 12 (3), 3241. 2nd edn. Jossey-Bass Publishing, San Francisco, CA.
Flynn L., Carryer J. & Budge C. (2005) Organizational attri- Kovner C. & Gergen P. (1998) Nurse staffing levels and adverse
butes valued by hospital, home care, and district nurses in the events following surgery in U.S. hospitals. Image: Journal of
United States and New Zealand. Journal of Nursing Scholar- Nursing Scholarship 30 (4), 315321.
ship 37, 6772. *Kroposki M. & Alexander J.W. (2006) Correlation among
*Flynn L., Liang Y., Dickson G.L. & Aiken L.H. (2010) Effects client satisfaction, nursing perception of outcomes, and
of nursing practice environments on quality outcomes in nurs- organizational variables. Home Healthcare Nurse 24 (2),
ing homes. Journal of the American Geriatrics Society 58 8794.
(12), 24012406. Lake E.T. (2002) Development of the practice environment
*Gardner J.K., Thomas-Hawkins C., Fogg L. & Latham C.E. scale of the Nursing Work Index. Research in Nursing and
(2007) The relationships between nurses perceptions of the Health 25 (3), 176188.
hemodialysis unit work environment and nurse turnover, Lake E.T. & Friese C.R. (2006) Variations in nursing practice
patient satisfaction, and hospitalizations. Nephrology Nursing environments: relation to staffing and hospital characteristics.
Journal 34 (3), 271281. Nursing Research 55, 19.
Gilmartin M. & DAunno T.A. (2007) Chapter 8: Leadership Larrabee J.H., Engle V.F. & Tolley E.A. (1995) Predictors of
research in healthcare. The Academy of Management Annals patient-perceived quality. Scandinavian Journal of Caring Sci-
1 (1), 387438. ences 9, 153164.
Ginsburg L., Norton P.G., Casebeer A. & Lewis S. (2005) An *Larrabee J.H., Ostrow C.L., Withrow M.L., Janney M.A.,
educational intervention to enhance nurse leaders perceptions Hobbs G.R. & Buran C. (2004) Predictors of patient satisfac-
of patient safety culture. Health Services Research 40 (4), tion with inpatient hospital nursing care. Research in Nursing
9971020. and Health 27, 254268.
Goleman D., Boyatzis R. & McKee A. (2002) The New Lead- Laschinger H.K.S., Purdy N. & Almost J. (2007) The impact of
ers: Transforming the Art of Leadership into the Science of leadermember exchange quality, empowerment, and core
Results. Little Brown, London, UK. self-evaluation on nurse managers job satisfaction. The Jour-
Graen G. & Uhl-Bien M. (1995) Relationship-based approach nal of Nursing Administration 37 (5), 221229.
to leadership: development of leader-member exchange Lee H. & Cummings G.G. (2008) Examining the relationship
(LMX) theory of leadership over 25 years: applying a between director leadership practices and manager worklife
multi-level multi-domain perspective. Leadership Quarterly 6, and burnout. Journal of Leadership Studies 2 (2), 4762.
219247. Lowe G.S. (2005) Raising the bar for people practices: helping
Hage J. & Aiken M. (1969) Routine technology, social struc- all health organizations become preferred employers. Health-
ture, and organization goals. Administrative Science Quar- care Quarterly 8 (1), 6063.
terly 14, 366376. MacPhee M., Wardrop A. & Campbell C. (2010) Transforming
*Hansen L.O., Williams M.V. & Singer S.J. (2011) Perceptions work place relationships through shared decision making.
of hospital safety climate and incidence of readmission. Journal of Nursing Management 18, 10161026.
Health Services Research 46 (2), 596616. Mark B.A., Sayler J. & Smith C.S. (1996) A theoretical model
*Havig A.K., Skogstad A., Kjekshus L.E. & Romoren T.I. for nursing systems outcomes research. Nursing Administra-
(2011). Leadership, staffing and quality of care in nursing tion Quarterly 20 (4), 1227.
homes. BMC Health Services Research 11 (327), doi:10.1186/ Mark B.A., Hughes L.C. & Jones C.B. (2004) The role of the-
1472-6963-11-327. ory in improving patient safety and quality health care. Nurs-
*Houser J. (2003) A model for evaluating the context of nursing ing Outlook 52 (1), 1116.
care delivery. Journal of Nursing Administration 33 (1), 39 McClelland D.C. (1987) Human Motivation. Cambridge Uni-
47. versity Press, Cambridge, MA.

2013 John Wiley & Sons Ltd


722 Journal of Nursing Management, 2013, 21, 709724
Systematic review update

McCutcheon A.S., Doran D., Evans M., McGillis-Hall L. & Shortell S.M. & Kaluzny A.D. (2000) Health Care Manage-
Pringle D. (2009) Understanding nurse manager stress and ment: Organization Design and Behavior, 4th edn. Delmar
work complexity: factors that make a difference. Nursing Publishers, Albany, NY.
Leadership 22 (3), 4867. Shortell S.M., Rousseau D.M., Gillies R.R., Devers K.J. &
McGillis Hall L., Doran D., Baker R.G. et al. (2003) Nurse Simons T.L. (1991) Organizational assessment in intensive
staffing models as predictors of patient outcomes. Medical care units (ICUs): construct development, reliability, and
Care 41 (9), 10961109. validity of the ICU nurse-physician questionnaire. Medical
McGillis-Hall L. & Doran D. (2007) Nurses perceptions of Care 29 (8), 709726.
hospital work environment. Journal of Nursing Management Singer S., Lin S., Falwell A., Gaba D. & Baker L. (2009) Rela-
15, 264273. tionship of safety climate and safety performance in hospitals.
*McNeese-Smith D.K. (1999) The relationship between mana- Health Services Research 44 (2), 399421.
gerial motivation, leadership, nurse outcomes and patient Stahl M.J. (1986) Power Motivation. Praeger, New York, NY.
satisfaction. Journal of Organizational Behavior 20, 243 *Taylor J.A., Dominici F., Agnew J., Gerwin D., Morlock L. &
259. Miller M.R. (2012) Do nurse and patient injuries share com-
Meterko M., Nelson E.C. & Rubin H.R. (1990) Patient judg- mon antecedents? An analysis of associations with safety
ments of hospital quality: report of a pilot study. Medical climate and working conditions. BMJ Quality and Safety 21
Care 28, S1S56. (2), 101111.
Mitchell P. & Shortell S.M. (1997) Adverse outcomes and vari- Thomas-Hawkins C., Denno M., Currier H. & Wick G. (2003)
ations in organization of care delivery. Medical Care 35, Staff nurses perceptions of the work environment in free-
NS19NS32. standing hemodialysis facilities. Nephrology Nursing Journal
Needleman J., Buerhaus P., Mattke S., Stewart M. & Zelevin- 30 (2), 169183.
sky K. (2002) Nurse-staffing levels and the quality of care in Thompson D.N., Hoffman L.A., Sereika S.M. et al. (2011) A
hospitals. New England Journal of Medicine 346 (22), 1715 relational leadership perspective on unit-level safety climate.
1722. Journal of Nursing Administration 41 (11), 479487.
Northouse P.G. (2001) Leadership: Theory and Practice. Sage, Tourangeau A.E. (2005) A theoretical model of the determinants
Thousand Oaks, CA. of mortality. Advances in Nursing Science 28 (1), 58Y69.
Page A. (2004) Keeping Patients Safe Transforming the Work Tourangeau A.E., Giovannetti P., Tu J.V. & Wood M. (2002)
Environment of Nurses. National Academies Press, Washing- Nursing-related determinants of 30-day mortality for hospi-
ton, DC. talized patients. Canadian Journal of Nursing Research 33
*Paquet M., Courcy F., Lavoie-Tremblay M., Gagnon S. & (4), 7188.
Maillet S. (2013) Psychosocial work environment and pre- *Tourangeau A.E., Doran D.M., McGillis Hall L. et al. (2007)
diction of quality of care indicators in one Canadian health Impact of hospital nursing care on 30-day mortality for
center. Worldviews on Evidence-Based Nursing 10 (2), 82 acute medical patients. Journal of Advanced Nursing 57 (1),
94. 3244.
Peterson S.J., Walumbwa F.O., Avolio B.J. & Hannah S.T. Tregunno D., Jeffs L., McGillis Hall L., Baker R., Doran D. &
(2012) The relationship between authentic leadership and fol- Bassett S. (2009) On the ball leadership for patient safety
lower job performance: the mediating role of follower positiv- and learning in critical care. Journal of Nursing Administra-
ity in extreme contexts. The Leadership Quarterly 23 (3), tion 39 (7/8), 334339.
502516. Vogus T.J. & Sutcliffe K.M. (2007a) The Safety Organizing
Polit D. & Beck C. (2011) Nursing Research: Generating and Scale: development and validation of a behavioral measure of
Assessing Evidence for Nursing Practice, 9th edn. Lippincott safety culture in hospital nursing units. Medical Care 45 (1),
Williams & Wilkins, Philadelphia, PA. 4654.
*Pollack M.M. & Koch M.A. (2003) Association of outcomes *Vogus T.J. & Sutcliffe K.M. (2007b) The impact of safety
with organizational characteristics of neonatal intensive care organizing, trusted leadership, and care pathways on reported
units. Critical Care Medicine 31 (6), 16201629. medication errors in hospital nursing units. Medical Care 45
*Raup G.H. (2008) The impact of ED nurse manager leadership (10), 9971002.
style on staff nurse turnover and patient satisfaction in aca- Weick K.E. & Sutcliffe K.M. (2001) Managing the Unexpected:
demic health center hospitals. Journal of Emergency Nursing Assuring High Performance in an Age of Complexity. Jossey-
34 (5), 403409. Bass, San Francisco, CA.
Roberts K. & Reilly C. (1974) Measuring organizational com- Westra B.L., Cullen L., Brody D., Jump P., Geanon L. & Miland
munication. Journal of Applied Psychology 59, 321326. E. (1995) Development of the home care client satisfaction
Rubin H., Ware J.E. & Hayes R.D. (1990) Patient judgments of instrument. Public Health Nursing 12 (6), 393399.
hospital quality questionnaire. Medical Care 28 (9), 2243. Wong C.A. & Cummings G.G. (2007) The relationship between
Sexton J.B., Thomas E.J. & Helmreich R.L. (2000) Error, stress, nursing leadership and patient outcomes: a systematic review.
and teamwork in medicine and aviation: cross sectional Journal of Nursing Management 15 (5), 508521.
surveys. British Medical Journal 320 (7237), 745749. Wong C.A. & Cummings G.G. (2009) Authentic leadership: a
Sheridan J.E., White J. & Fairchild T.J. (1992) Ineffective staff, new theory for nursing or back to basics? Journal of Health
ineffective supervision, or ineffective administration? Why Organization and Management 23 (5), 522538.
some nursing homes fail to provide adequate care. The Wong C.A., Laschinger H.K., Cummings G.G., Vincent L. &
Gerontologist 32 (3), 334341. OConnor P. (2010) Decisional involvement of senior nurse

2013 John Wiley & Sons Ltd


Journal of Nursing Management, 2013, 21, 709724 723
C. A. Wong et al.

leaders in Canadian acute care hospitals. Journal of Nursing Yukl G.A. (2006) Leadership in Organizations. Pearson Prentice
Management 18 (2), 122133. Hall, Upper Saddle River, NJ.
Xin K.R. & Pelled L.H. (2003) Supervisor-subordinate conflict
and perceptions of leadership behaviour: a field study. The
Supporting information
Leadership Quarterly 14 (1), 2540.
Young-Ritchie C., Laschinger H.K.S. & Wong C. (2009) The Additional Supporting Information may be found in
effects of emotionally intelligent leadership behaviour on
the online version of this article:
emergency staff nurses workplace empowerment and organi-
zational commitment. Nursing Leadership 22 (1), 7085. Appendix S1. Characteristics of included studies.

2013 John Wiley & Sons Ltd


724 Journal of Nursing Management, 2013, 21, 709724