Nurse Education Today 62 (2018) 91–97

Contents lists available at ScienceDirect

Nurse Education Today

journal homepage: www.elsevier.com/locate/nedt

Turning education into action: Impact of a collective social education T

approach to improve nurses' ability to recognize and accurately assess
delirium in hospitalized older patients
⁎
Catherine Traversa, , Amanda Hendersonb, Fred Grahamc, Elizabeth Beattied
a
Dementia Collaborative Research Centre, School of Nursing, Queensland University of Technology, Victoria Park Road, Kelvin Grove Q 4059, Australia
b
Nursing Practice Development Unit, Princess Alexandra Hospital, 199 Ipswich Rd, Woolloongabba, Qld 4102, Australia
c
Dementia and Delirium, Princess Alexandra Hospital, 199 Ipswich Rd, Woolloongabba, Qld 4102, Australia
d
Dementia Centre for Collaborative (DCRC School of Nursing), Queensland University of Technology (QUT), Level 6, N Block, Victoria Park Rd., Kelvin Grove, Qld 4059,
Australia

A R T I C L E I N F O A B S T R A C T

Keywords: Background: Although cognitive impairment including dementia and delirium is common in older hospital pa-
Dementia tients, it is not well recognized or managed by hospital staff, potentially resulting in adverse events. This paper
Delirium describes, and reports on the impact of a collective social education approach to improving both nurses'
Hospitals knowledge of, and screening for delirium.
Education
Methods: Thirty-four experienced nurses from six hospital wards, became Cognition Champions (CogChamps) to
Nursing
lead their wards in a collective social education process about cognitive impairment and the assessment of
delirium. At the outset, the CogChamps were provided with comprehensive education about dementia and
delirium from a multidisciplinary team of clinicians. Their knowledge was assessed to ascertain they had the
requisite understanding to engage in education as a collective social process, namely, with each other and their
local teams. Following this, they developed ward specific Action Plans in collaboration with their teams aimed at
educating and evaluating ward nurses' ability to accurately assess and care for patients for delirium. The plans
were implemented over five months. The broader nursing teams' knowledge was assessed, together with their
ability to accurately assess patients for delirium.
Results: Each ward implemented their Action Plan to varying degrees and key achievements included the
education of a majority of ward nurses about delirium and the certification of the majority as competent to assess
patients for delirium using the Confusion Assessment Method. Two wards collected pre-and post-audit data that
demonstrated a substantial improvement in delirium screening rates.
Conclusion: The education process led by CogChamps and supported by educators and clinical experts provides
an example of successfully educating nurses about delirium and improving screening rates of patients for de-
lirium.
Trial Registration: ACTRN 12617000563369.

1. Background
Cognitive impairment (CI) including dementia and delirium is
common in older patients admitted to acute care hospitals and patients
with these conditions face a disproportionate risk of preventable harm
in this environment (Inouye et al., 2014a). Thus, improving the care of
these patients when hospitalized and their outcomes has been re-
cognized as a priority issue by leading healthcare organizations in-
cluding Australia's Commission on Safety and Quality in Healthcare
(ACSQHC, 2016). A key issue for ACSQHC is improving staff knowledge
that can inform care practices so CI can be identified early. Currently,
both dementia and delirium are poorly recognized in hospitals and
infrequently documented (Australian Institute of Health and Welfare
(AIHW);, 2013; Cummings et al., 2011; Rice et al., 2011), with reports
indicating that dementia is not recorded in almost half of all admissions
(47% not recorded; Australian Institute of Health and Welfare (AIHW),
2013), while delirium is missed in as many as 75% of cases (Rice et al.,
2011). Reasons include the lack of assessment of cognitive functioning,

⁎
Corresponding author.
E-mail addresses: catherine.travers@qut.edu.au (C. Travers), Amanda.Henderson@health.qld.gov.au (A. Henderson), Frederick.Graham@health.qld.gov.au (F. Graham),
Elizabeth.Beattie@qut.edu.au (E. Beattie).

https://doi.org/10.1016/j.nedt.2017.12.026
Received 23 July 2017; Received in revised form 20 November 2017; Accepted 20 December 2017
0260-6917/ © 2018 Elsevier Ltd. All rights reserved.
C. Travers et al.
a lack of knowledge amongst staff including poor understanding of the
benefits of the early recognition and treatment of delirium, and in
distinguishing between these and other conditions that may have si-
milar presentations (e.g. stroke, poly-pharmacy issues affecting cogni-
tion, alcohol withdrawal, (Australian Institute of Health and Welfare
(AIHW), 2013, p9; Phillips et al., 2011; Teodorczuk et al., 2012).
Failure to recognize CI early in the admission means these patients are
vulnerable to adverse outcomes including delirium, therefore ensuring
nurses are knowledgeable about delirium is essential (Inouye et al.,
2014a).
The under-detection of both dementia and delirium (Cummings
et al., 2011; Rice et al., 2011), persists despite the availability of several
brief screening tools to assess both dementia and delirium in the hos-
pital environment (Jackson et al., 2013; Wong et al., 2010). This is
partly because hospital nurses have not received adequate education
about dementia and delirium and how to recognize CI (Gandesha et al.,
2012; Gladman et al., 2012; Hynninen et al., 2015). While there is a
need to educate hospital nurses about recognizing dementia and de-
lirium so care can be appropriately individualized, there is limited
evidence regarding the most effective ways of doing so.
The simple provision of guidelines and education of staff about
delirium detection is generally insufficient to result in practice change
(Wand, 2011; Young and George, 2003), indicating the need for a dif-
ferent approach. One approach that has been shown to be conducive to
learning and improvement in schools is ‘Distributed Leadership’ (DL;
Bolden, 2011), and while it has been predominantly studied in the field
of school education, DL is increasingly being studied within healthcare.
DL can be conceived of as a collective social process that involves in-
dividuals pooling and sharing their skills and expertise such that the
result is greater than the sum of their parts (Bolden, 2011, p252). Thus,
in this project, the approach adopted to educate hospital nurses about
dementia and delirium and increase screening for delirium early in
patients' admissions was a collective social process involving academics
(the Research Team), nurse consultants, and the nursing teams deli-
vering care. Responsibility for education was distributed across the
different levels of clinical nurses, and the nurses working with their
specific teams, thereby engendering a collective responsibility for
nursing staff to engage with the learning (Ritchie and Woods, 2007).
The approach also recognizes that nurses prefer to access evidence
based knowledge socially by engaging with local clinical experts rather
than with online guidelines and text (Estabrooks et al., 2005; Marshall
et al., 2011). In this project the Cognitive Champions took the leading
role, assumed responsibility, and responded specifically to emerging
needs in their wards to tailor and organize learning with their collea-
gues to ensure the provision of delirium education was timely and re-
levant. This use of a collective social process as a medium to educate
nurses was part of a larger study that targeted education and practice
change aimed at increasing the uptake of best practices for delirium
prevention, management and treatment (Travers et al., 2017).
This paper reports on the processes surrounding the educational
component of the intervention only and the impact on nurses' knowl-
edge. Other aspects of the intervention including patient outcomes and
qualitative evidence regarding barriers and factors that facilitated the
project's implementation are discussed in greater detail in a companion
article.
2. Methods
The effectiveness of the collective social education process and its
impact on nurses' knowledge of delirium was evaluated by:
- the percentage of staff that engaged in the in-service educational
sessions about delirium,
- increased nurses' performance on knowledge tests for the accurate
identification of delirium, and
- improvements in timely and accurate assessments of at-risk patients
92
Nurse Education Today 62 (2018) 91–97
for delirium.
2.1. Distributed Leadership Approach
Experienced nurses (with more than two years clinical experience),
and (a) a specific interest in CI, or (b) leadership skills, were identified
to become Cognition Champions (CogChamps). They were provided
with comprehensive dementia and delirium education and training
(Workshop One) and the distributed leadership approach to guide in-
service sessions in their clinical areas (Workshop Two), through two full
days of Workshops conducted six weeks apart. An important component
of Workshop One was the assessment of delirium using the short
Confusion Assessment Method (CAM; Inouye et al., 1990, 2014b). The
session included a demonstration of the CAM by the hospital's dementia
and delirium specialist (FG: Clinical Nurse Consultant — Dementia and
delirium), practice using the CAM in pairs and supervised live practice
by the CogChamps on their home ward. This was consolidated by a
second live observation by an expert (CT) of the CogChamps per-
forming a CAM assessment on their ward. This occurred in the month
following Workshop One and CogChamps were assessed by an expert
clinician/researcher for their ability to both administer and interpret
the CAM. If necessary, the observation was repeated until the Cog-
Champ was deemed to be competent in both domains. Establishing the
lead nurses' ability to perform these assessments accurately was es-
sential to ensuring that they had the requisite knowledge and under-
standing of delirium. The collective social process (DL approach) was
based on the development of ward specific Action Plans led by the
CogChamps, thus ensuring the action items reflected each ward's
priorities and requirements. Each plan included specific actions for
nurse education with measurable outcomes and associated timeframes.
Initially, there were approximately six CogChamps for each of six
wards. One or more CogChamp(s) from each ward met (face to face
and/email) on a regular basis (weekly) to refine the education in-
itiatives in collaboration with the Research team. The research team
also supported the CogChamps throughout the five month im-
plementation by assisting them with providing education sessions,
sourcing resources, providing feedback and providing a data analytic
service to assist the CogChamps analyze any data they collected.
The collective social education process also involved the nomination
of a CogChamp as leader for each ward, the development of sound
communication processes to communicate amongst themselves about
the project (e.g. use of a communications book and email to schedule
meetings and send updates), and the education activities. The project
was also included as an agenda item at their regular ward meetings to
involve all staff in the progress, and to address issues of sustainability.
Educational resources to support the efforts (e.g. leaflets, lanyards with
the CAM printed, and posters) were also sourced and distributed to
nurses, and posters were displayed in prominent locations around the
ward. The actions implemented by each ward are summarized in
Table 1.
2.2. Setting
The education process was conducted across six wards (4 medical, 2
surgical) of a large tertiary referral hospital in South-East Queensland,
Australia. The education initiative had the support of the hospital ex-
ecutive and key hospital staff including the Nurse Unit Managers of the
wards.
3. Ethics
Ethics approvals were obtained from the relevant Human Research
Ethics Committees responsible for the participating hospital. As no
personally identifying information was recorded by the nurses (they
were asked to use a personally meaningful alias on questionnaires for
identification purposes), obtaining individual informed consent from
C. Travers et al. Nurse Education Today 62 (2018) 91–97

Table 1
Summary of social collective education activities across the six wards.

Area Action# A B C D E F
Procedural Nomination of a CogChamp leader for each ward

Development of a ward specific Action Plan

Communication Establishment of sound communication processes to enable

the CogChamps to communicate amongst themselves about
the project

Inclusion of the CogChamps project as an agenda item at

regular ward meetings

Knowledge Delirium education sessions for nurses and bed-side

education
Assessment of nurse’s delirium knowledge pre-education

Assessment of nurse’s delirium knowledge post-education

Assessment of nurses’ proficiency in assessing patients for

delirium using the CAM

Distribution of resources (e.g. posters, information sheets) to

support educational activities

Increasing the Audit of CAM assessments pre-and post education

assessment of
patients for
delirium

#
Shading indicates activity not undertaken.

each participant was not required by the Ethics Committees. 3.3. Summary of CogChamps' Knowledge, Understanding and Confidence

3.1. Baseline Activity to Assure Knowledge, Understanding and Confidence
of CogChamps to Lead the Collective Social Education Process
3.1.1. Measures
CogChamps were assessed for their knowledge and self-rated con-
fidence and stress when nursing patients with dementia, confusion or
agitation. A questionnaire administered at both Workshop One (base-
line) and at the Commencement of Workshop Two determined baseline
information and subsequent changes. The questionnaire included (a)
one single item question assessing self-rated dementia knowledge using
a 1–5 scale where 1 = No/Very little knowledge and 5 = Very
knowledgeable; (b) 15 True/False items relating to delirium features
and risk factors (see Travers et al., 2017 for complete details), and (c)
five validated vignettes depicting a variety of patient presentations with
features of dementia and delirium which nurses were asked to identify
(McCrow, 2012). Another five vignettes matched for diagnostic com-
plexity were used on the second occasion. Nurses' self-confidence in
recognizing when a patient had delirium, and when nursing a patient
who appears to have dementia or is confused or agitated was assessed
using two single-item statements where 1 = not at all confident and
5 = very confident. Self-reported stress when nursing a patient with
delirium was also assessed using a similarly styled question.
3.2. Data Analysis
Demographic characteristics of the CogChamps are reported as de-
scriptive data only. Group differences and outcomes were assessed
using the Student's t-test for continuous data, and non-parametric tests
(Chi-squared statistic; Wilcoxon signed rank test) for categorical data.
All analyses were performed using SPSS for Windows version 21.0 (IBM
Corp, 2012).
3.3.1. Demographic Data — CogChamps
CogChamps (n = 34) across the six wards participated in the in-
tervention; of those, 27 attended both Workshops, while seven attended
one Workshop only. Demographic data for 31 participants were avail-
able and results showed that the majority of the CogChamps were fe-
male (84%), aged between 21 and 40 years (84%), and had worked as a
nurse for three or more years (71%). Over three-quarters (77.4%) re-
ported that they had received some education regarding dementia or
delirium through their workplace, although 20% reported having re-
ceived no such education.
3.4. Delirium Knowledge — CogChamps
At baseline (immediately prior to Workshop One), 31 CogChamps
(94%) completed a baseline knowledge questionnaire with most rating
their dementia knowledge as fair or greater (67.7%), which is sub-
stantiated by their performance on the objective measures. Overall,
they achieved high scores on the 15 True/False items, and demon-
strated a high level of ability to correctly identify dementia and/or
delirium as depicted in the vignettes (see Table 2). Thus, the Cog-
Champs had a sound knowledge base regarding dementia and delirium,
essential for their role as educators in this project. Table 2 also shows a
significant increase in CogChamps' self-rated dementia knowledge and
self-confidence in recognizing delirium following Workshop One (28
CogChamps completed a post-Workshop questionnaire) with no other
changes evident.
3.5. CAM Competency Assessments — CogChamps
Thirty-one CogChamps were observed performing a live patient
assessment using the CAM in the month following Workshop One by

93
C. Travers et al. Nurse Education Today 62 (2018) 91–97

Table 2
Pre- versus post-Workshop One responses and significance of change.

Self-report items Response P-value

No/Very little A little n (%) A Fair amount/Somewhat n (%) Quite a bit/Moderately Very n (%)
n (%) n (%)

Self-rated dementia knowledge

Pre-Workshop One (n = 31) 0 10 (32.2) 14 (45.2) 6 (19.4) 1 (3.2) 0.003a
Post-Workshop One (n = 28) 0 5 (17.8) 8 (28.6) 15 (53.6) 0

Self-confidence in recognizing delirium

Pre-Workshop 0 3 (9.7) 13 (41.9) 13 (41.9) 2 (6.5) 0.008a
Post-Workshop 0 1 (3.6) 7 (25.0) 19 (67.8) 1 (3.6)

Self-confidence when nursing someone with dementia/

delirium
Pre-Workshop 0 1 (3.2) 10 (32.3) 17 (54.8) 3 (9.7) 0.17
Post-Workshop 0 1 (3.6) 7 (25.0) 19 (67.8) 1 (3.6)

Self-rated stress when nursing a patient with delirium

Pre-Workshop 3 (10.0) 15 (50.0) 10 (33.3) 2 (6.7) 0 0.44
Post-Workshop 3 (10.7) 17 (60.7) 5 (17.9) 3 (10.7) 0

Knowledge assessment Total correct (average) Range p-Value

True/False Questions (15)

Pre-Workshop 13.90; SDb = 0.98 12–15 0.33
Post-Workshop 14.04; SD = 1.14 12–15

Vignettes Correct n (%) Incorrect n (%) p-Value

Identification of dementia
Pre-Workshop 29 (93.5) 2 (6.5) 0.66
Post-Workshop 25 (89.3) 3 (10.7)
Identification of delirium
Pre-Workshop 26 (83.9) 5 (16.1) 0.32
Post-Workshop 25 (89.3) 3 (10.7)
Identification of delirium superimposed on dementia
Pre-Workshop 26 (83.9) 5 (16.1) 0.56
Post-Workshop 24 (85.7) 4 (14.3)

a
p < 0.05.
b
SD = Standard deviation.

CT. All were certified as being competent to both administer and in-
terpret the CAM although two CogChamps were required to repeat the
CAM assessment prior to receiving certification.
3.6. Attrition
Although the project commenced with 34 CogChamps, attrition was
an issue, primarily due to staff being re-assigned to other projects
(n = 2), resigning from the hospital (n = 1), going on extended leave
(n = 1), or being too busy with other work to be actively involved in
CogChamps (n = 2). Thus, by October 2016, 28 CogChamps remained
actively involved in the project.
4. Collective Social Education Activity
4.1. Education about Cognitive Impairment and Accurate Patient
Assessment
The collective social education activity varied across the six wards
with most progress occurring on the ward with the highest level of
Nurse Unit Manager (NUM) involvement (Ward A). The support of the
NUMs was essential for the CogChamps' attendance at the Workshops
and allowing them to devote time to educational activities.
Education sessions were designed to accommodate nurses' rosters
and address the time constraints of the acute hospital environment and
hence were brief (around 40 min) and were provided at the beginning/
end of shifts on multiple days to facilitate nurses' attendance. The ses-
sions were supplemented by one-to-one bedside coaching by the
CogChamps and who also provided instruction in entering the results of
the CAM assessments onto the hospital's electronic medical records
system. The CogChamps maintained records of attendance. All wards
aimed to educate 80% of nurses. Following this, the CogChamps ob-
served nurses assessing a patient for delirium using the CAM. Nurses
were required to demonstrate competency in both administering and
interpreting the CAM before being awarded a certificate of competency.
Nurses received continuing practice development points for both ac-
tivities.
5. Results
Seven delirium education sessions were delivered by FG who also
trained a number of the CogChamps to present the standardized session.
Over the implementation period, 148 ward nurses attended a delirium
education session supplemented by bed-side coaching and of those, 118
nurses were observed by a CogChamp assessing a patient for delirium
using the CAM and were certified as being proficient in using the tool.
Thus, by the end of October, three of the six wards successfully
achieved the targeted 80% of nurses receiving education about CI,
while one ward achieved the targeted 80% of nurses certified as com-
petent to assess delirium using the CAM (see Table 3).
CogChamps from three wards collected delirium knowledge data
(using the vignettes provided by the Research team) from nurses prior
to the educational sessions and the results revealed a number of
knowledge deficits. The results are presented in Table 3 and nurses on
all wards were able to identify dementia more accurately than delirium
or delirium superimposed on dementia. By comparison, the CogChamps
were considerably more accurate in identifying the latter two condi-
tions (see Table 2). Only CogChamps from two wards were able to re-

94
C. Travers et al. Nurse Education Today 62 (2018) 91–97

Table 3
Initiatives implemented by CogChamps on the intervention wards.

Area of change Ward

Delirium knowledge A B C D E F Total

Nurses who attended a delirium education session 32 (89)a 21 (48) 30 (81)a 25 (89)a 20 (71) 20 (50) 148 (69.5)
n (% of total number of ward nurses)
a
Nurses certified as competent to assess a patient for delirium using the CAM 32 (89) 10 (23) 20 (54) 21 (75) 19 (67) 16 (40) 118 (77)
n (% of total number of ward nurses)
Assessment of nurses' knowledge pre-education
Number of assessments completed 23 36 22
Results — Vignettes:
Correct identification of Dementia 78% 94% 96%
Correct identification of Delirium 72% 78% 57%
Correct identification of Delirium superimposed on Dementia 74% 75% 68%
True/False Items — Average, (SD) 13.52 (1.7) 13.53 (1.6)
Range 7–15 8–15
Assessment of nurses' knowledge post-educationb
Number of assessments completed
Results — Vignettes: 12
Correct identification of Dementia 92%
Correct identification of Delirium 83%
Correct identification of Delirium superimposed on Dementia 71%

Increased delirium assessments using the CAM — audit data

Audit of CAM assessments — Pre-Intervention
Number of patients audited 72 12
Number of patients who should have been assessed using a CAM (% of number audited) 14 (19) 3 (25)
Number of patients for whom a CAM was performed (% of at-risk patients) 2 (14) 2 (17)
Number of correctly completed CAMs (% of CAMs completed) 1 (7) 0
Audit of CAM assessments — Post-Intervention
Number of patients audited 5810 (17) 12
Number of patients who should have been assessed using a CAM (% of number audited) 6 (60) 8 (67)
Number of patients for whom a CAM was performed (% of at-risk patients) 6 (60) 3 (38)
Number of correctly completed CAMs (%) 2 (25)

a
Target of 80% achieved
b
Only three of the pre- and post-assessments were completed by the same nurse.

administer the vignettes following the educational component (re-
quiring much persistence), and the results from one ward are reported
in Table 3 (results from the second were unavailable). Numbers are
small and were completed by different nurses on the two occasions,
with three exceptions, and hence it is not possible to draw any con-
clusions regarding the effectiveness of the education.
5.1. Increasing Assessments of ‘at-Risk’ Patients for Delirium
While increasing the numbers of ‘at-risk’ patients assessed for de-
lirium within 24 h of their admission to the ward was included as an
Action item on all plans, CogChamps from two wards also included
objective measures of success (target = 80%). This involved con-
ducting mini-audits before and after the delirium education program
using a brief audit tool they were assisted to develop by the Research
Team (FG, CT), and a checklist to accurately identify patients at risk,
and who therefore should have been assessed for delirium. The audits
were completed by the CogChamps for one day on both occasions and
the data were subsequently analyzed by a Research Team member (CT)
who provided a brief summary of the results to the CogChamps. The
results of the two audits of CAM assessments undertaken by the
CogChamps showed that delirium assessments were infrequently con-
ducted, or were reported incorrectly prior to the delirium education
component and CAM training (July 2016) — see Table 3. By compar-
ison, data collected in October showed substantial increases in the
proportion of CAMs performed, and importantly in the proportion
correctly reported.
6. Discussion
This paper demonstrates the application of a Distributed Leadership
model within the acute hospital environment to educate hospital nurses
about CI and improving screening of at-risk patients for delirium early
in their admission. The collective social approach harnessed the capa-
city of Cognition Champions (CogChamps), the research team and the
NUMs to work collaboratively to disseminate education and upskill
hospital nurses across six wards. Key achievements included the edu-
cation of a majority of ward nurses about dementia and delirium and
the upskilling of the majority to become competent to assess patients
for delirium using the CAM. Evidence that this education was translated
into practice is found in the results of the mini-audits of CAM usage
undertaken in two wards. In both instances, the post-audit results
showed a significant increase in screening rates compared to the
baseline audit. Other achievements included the development or ac-
quisition of resources to support nurses' learning.
The CogChamps demonstrated a sound knowledge of dementia and
delirium at commencement, confirming they had the requisite level of
knowledge for their role as educators and mentors. Importantly, their
self-rated dementia knowledge and self-confidence in recognizing de-
lirium increased significantly following Workshop One, which is likely
to have improved their confidence to lead change on their wards —
essential for the leadership role (Rankin et al., 2016).
The delirium education included brief education sessions delivered
on multiple occasions, supplemented by bed-side coaching to improve
nurses' competency to assess patients for delirium, and the adoption of
resources to support learning. Throughout, the CogChamps received
support from both the NUMs and the Research Team who assisted them
by providing education, sourcing resources and providing feedback
regarding progress. An important element of the intervention was the
development of ward specific Action Plans, led by the CogChamps.
They were responsible for identifying the key actions to be undertaken,
thus tailoring each plan to each ward's specific requirements and cir-
cumstances. Acknowledging the need for change has been identified as
important in motivating implementation efforts (Dogherty et al., 2013),

95
C. Travers et al.
and the involvement of the CogChamps in this process appears to have
engaged them sufficiently to lead change efforts on their wards. In this
project, the Action Plans provided a concrete guide and clear direction
for the educational efforts, not only for the CogChamps but also for the
Research team and others involved which may have fostered a united
team effort.
This is the first attempt, of which we are aware, to apply the concept
of DL to educate hospital nurses about dementia and delirium and to
improve delirium screening rates. The model successfully overcame the
usual challenges faced in this environment including busyness, time
constraints, completing priorities, shift-work and attrition.
The implementation of this project coincided with the identification
of the care of hospitalized patients with CI as a priority issue in
Australia — so much so that a national Cognitive Care campaign was
launched in 2016 (Australian Commission on Safety and Quality in
Health Care (ACSQHC), 2016), to which the hospital made a public
commitment (they are publicly acknowledged on the campaign web-
site). Thus, the timing of the project was fortuitous. The project was
also implemented in an environment in which a sound foundation for
caring for patients with CI had been established through the prior im-
plementation of several initiatives by FG (see Travers et al., 2017).
Although these circumstances provided a favorable environment con-
ducive to the project's success, it was nevertheless, time intensive and
required a great deal of dedicated effort for all involved, highlighting
the amount of effort required in effective practice change in a complex
environment.
6.1. Study Strengths and Weaknesses
A strength of the study was its conduct in a real-world hospital
setting with its attendant challenges, and shows how the existing nur-
sing workforce can be effectively supported to work together to dis-
seminate knowledge and skills. It utilizes the teamwork approach used
in hospitals and hence could be readily adopted by other hospitals.
While one limitation may be our reliance on data collected by the
CogChamps as evidence of the translation of knowledge into practice,
and it is acknowledged these data may not have been collected with as
much rigor as is ideal (e.g. the data collectors were not blinded), our
project was conducted in a real-world setting using real-world data.
Audit is a tool that many nurses are familiar with and is frequently used
for routine monitoring purposes at the hospital where our project was
conducted. Importantly, the CogChamps were able to complete the
audits within their rostered shifts and their involvement in the data
collection process is likely to have increased their engagement in the
project, particularly when they were able to see positive results from
their endeavors. Because the CogChamps were encouraged to collect
the measures of success, rather than the Research team doing this for
them, the down-side was that there are very limited data regarding the
impact of the educational component itself on nurse's knowledge.
One limitation of this project as reported in this manuscript includes
the failure to link the educational intervention and knowledge im-
provements to improved patient outcomes, and no data are reported
regarding the sustainability of the observed changes over the longer
term. Both of these short-comings will be addressed in a companion
manuscript (currently in preparation) in which the results of data col-
lected through direct observations of nurse: patient interactions and
chart audits, and patient outcome data (including 3 month follow-up
data) will be presented and discussed.
Finally, the project was implemented in one hospital only and its
transferability to other hospitals is not known, particularly hospitals
that do not have access to dementia specialist support and the resources
that supported the CogChamps project.
7. Conclusion
The CogChamps project demonstrates that nurse Champions can be
96
Nurse Education Today 62 (2018) 91–97
effectively empowered to educate other nurses about dementia and
delirium including the accurate recognition of delirium. A multifaceted
approach including engaging with the Cognition Champions
(CogChamps) to develop clear goals and the provision of a high level of
support and facilitation were central to implementing practice change.
Acknowledgements
This project was funded by an Aged Care Service Improvement and
Healthy Ageing Grant under the Department of Health and Ageing,
Australian Government's Ageing and Service Improvement Programme.
The efforts of the CogChamps in implementing this practice change
initiative is acknowledged with gratitude. This project and its successes
are attributable, in large part to their dedication and diligence.
References
Australian Commission on Safety and Quality in Health Care (ACSQHC). Caring for
Cognitive Impairment. Sydney: ACSQHC; 2016 ([cited 2016 Sept 10]. Available from:
https://www.safetyandquality.gov.au/our-work/cognitive-impairment/program-
overview/).
Australian Institute of Health and Welfare (AIHW). Dementia Care in Hospitals: Costs and
Strategies. Cat. no. AGE 72. Canberra: AIHW; 2013 ([cited 2016 Sept 16]. Available
from: http://www.aihw.gov.au/WorkArea/DownloadAsset.aspx?id=60129543386).
Bolden, R., 2011. Distributed leadership in organizations: a review of theory and re-
search. Int. J. Manag. Rev. 13, 251–269.
Cummings, E., Maher, R., Showell, C.M., Croft, T., Tolman, J., Vickers, J., et al., 2011.
Hospital coding of dementia: is it accurate? HIM J. 40, 5–11.
Dogherty, E.J., Harrison, M.B., Graham, I.D., Digel Vandyk, A., Keeping-Burke, L., 2013.
Turning knowledge into action at the point-of-care: the collective experience of
nurses facilitating the implementation of evidence-based practice. Worldviews Evid.-
Based Nurs. 10 (3), 129–139. http://dx.doi.org/10.1111/wvn.12009.
Estabrooks, C.A., Rutakumwa, W., O'Leary, K.A., et al., 2005. Sources of practice
knowledge among nurses. Qual. Health Res. 15 (4), 460–476.
Gandesha, A., Souza, R., Chaplin, R., Hood, C., 2012. Adequacy of training in dementia
care for acute hospital staff. Nurs. Older People 24 (4), 26–31.
Gladman JR, Porock D, Griffiths A, et al. 2012. Care of Older People with Cognitive
Impairment in General Hospitals. Nottingham: National Institute for Health Research;
(2012 [cited 2015 June 24]. Available from: http://www.netscc.ac.uk/hsdr/files/
project/SDO_FR_08-1809-227_V01.pdf).
Hynninen, N., Saarnio, R., Isola, A., 2015. The care of older people with dementia in
surgical wards from the point of view of the nursing staff and physicians. J. Clin.
Nurs. 24 (1–2), 192–201. http://dx.doi.org/10.1111/jocn.12669.
IBM Corp, 2012. IBM SPSS Statistics for Windows, Version 21.0. Armonk, NY, IBM Corp.
Inouye, S.K., van Dyck, C.H., Alessi, C.A., et al., 1990. Clarifying confusion: the confusion
assessment method. A new method for detection of delirium. Ann. Intern. Med. 113,
941–948.
Inouye, S.K., Westendorp, R.G., Saczynski, J.S., 2014a. Delirium in elderly people. Lancet
383, 911–922. http://dx.doi.org/10.1016/S0140-6736(13)60688-1.
Inouye, S.K., Kosar, C.M., Tommet, D., Schmitt, E.M., Puelle, M.R., Saczynski, J.S.,
Marcantonio, E.R., Jones, R.N., 2014b. The CAM-S: development and validation of a
new scoring system for delirium severity in 2 cohorts. Ann. Intern. Med. 160,
526–533. http://dx.doi.org/10.7326/M13-1927.
Jackson, T.A., Naqvi, S.H., Sheehan, B., 2013. Screening for dementia in general hospital
inpatients: a systematic review and meta-analysis of available instruments. Age
Ageing 42 (6), 689–695. http://dx.doi.org/10.1093/ageing/aft145.
Marshall, A.P., West, S.H., Aitken, L.M., 2011. Preferred information sources for clinical
decision making: critical care nurses' perceptions of information accessibility and
usefulness. Worldviews Evid.-Based Nurs. 8 (4), 224–235. http://dx.doi.org/10.
1111/j.1741-6787.2011.00221.x.
McCrow J. The Impact of a Model of Education to Improve Knowledge and Recognition of
Delirium in Older Persons by Registered Nurses (PhD Thesis). Brisbane, Qld:
Queensland University of Technology; 2012 ([cited 2015 Dec 9]. Available from:
http://eprints.qut.edu.au/55161/1/Judy_Maree_McCrow_Thesis.pdf).
Phillips, J., Pond, D., Goode, S.M., 2011. Timely Diagnosis of Dementia: Can we Do
Better? Alzheimer's Australia, Canberra.
Rankin, J., Matthews, A., Russell, M., Pollard, M., 2016. Clinical leadership and effective
nursing teams. J. Nurs. Health Care 4 (1), 76–81.
Rice, K.L., Bennett, M., Gomez, M., et al., 2011. Nurses' recognition of delirium in the
hospitalized older adult. Clin. Nurs. Spec. 25, 299–311. http://dx.doi.org/10.1097/
NUR.0b013e318234897b.
Ritchie, R., Woods, P.A., 2007. Degrees of distribution: towards an understanding of
variations in the nature of distributed leadership in schools. Sch. Leadersh. Manag. 27
(4), 363–381. http://dx.doi.org/10.1080/13632430701563130.
Teodorczuk, A., Reynish, E., Milisen, K., 2012. Improving recognition of delirium in
clinical practice: a call for action. BMC Geriatr. 12, 55. http://dx.doi.org/10.1186/
1471-2318-12-55.
Travers, C., Graham, F., Henderson, A., Beattie, E., 2017. CogChamps – a model of im-
plementing evidence-based care in hospitals: study protocol. BMC Health Serv. Res.
17, 202. http://dx.doi.org/10.1186/s12913-017-2136-0.
C. Travers et al. Nurse Education Today 62 (2018) 91–97

Wand, A.P., 2011. Evaluating the effectiveness of educational interventions to prevent delirium?: value of bedside instruments. JAMA 304 (7), 779–786. http://dx.doi.org/
delirium. Australas. J. Ageing 30, 175–185. http://dx.doi.org/10.1111/j.1741-6612. 10.1001/jama.2010.1182.
2010.00502.x. Young, L.J., George, J., 2003. Do guidelines improve the processes and outcomes of care
Wong, C.L., Holroyd-Leduc, J., Simel, D.L., Straus, S.E., 2010. Does this patient have in delirium? Age Ageing 32, 525–528.

97