IUPUI

Information at the Point

of Care
Development of a Mobile and Web-Based
Information Resource for Nurses
Christopher Kiess & Scott Russell
5/7/2009
Contents
Executive Summary .............................................................................................. 3
Introduction ........................................................................................................... 4
Fieldwork Activity & Data Collected ...................................................................... 5
Table 1: Literature Search ................................................................................. 5
Figure 1: Barriers to Information Seeking .......................................................... 7
Figure 2: Motivation to Seek Information ........................................................... 8
Goals & Requirements .......................................................................................... 8
Figure 3: Health Information Levels of Need ................................................... 10
Figure 4: Information Flow ............................................................................... 11
Salient Scenario 1 ........................................................................................... 11
Salient Scenario 2 ........................................................................................... 12
Salient Scenario 3 ........................................................................................... 13
Conceptual Design.............................................................................................. 14
Figure 5: First Conceptual Design ................................................................... 15
Figure 6: Final Conceptual Design .................................................................. 16
Walkthrough Scenario ..................................................................................... 18
Visual/Interactive Prototype ................................................................................ 19
Use Case 1: Asking a question ....................................................................... 20
Use Case 2: Getting a "live" answer................................................................ 26
Use Case 3: Giving an answer ........................................................................ 29
Preliminary User Feedback................................................................................. 33
Conclusion .......................................................................................................... 33
References: ........................................................................................................ 34
Appendix A ......................................................................................................... 35
Appendix B ......................................................................................................... 41
Appendix C ......................................................................................................... 55

2
Executive Summary

We embarked upon a project to remodel a popular SMS service and adapt it for the use of nurses
in a healthcare facility with the primary intention of giving them quicker, more efficient, mobile
access to information and knowledge-based resources. An intensive literature search was
conducted to identify nursing information needs. In addition we conducted a formal survey of 8
nurses in varying disciplines as well as conducted semi-structured interviews with nurses
concerning their current practice and information needs. Our conceptual design rested on the idea
that there should be human intervention at some level, but that we should be able to leverage
existing knowledge and return automated results using natural language processing and querying
the knowledge base. We developed a prototype in low-fidelity moving to a higher fidelity prototype
of a mobile-based and web-based application for seeking out information and developing a
knowledge-based resource for nursing in practice at the point of care. Our initial results suggest
this is an appropriate and feasible design from a conceptual standpoint. Our initial user test
indicates there is room and need for improvement of the model. Some primary considerations are
labeling of the interface, management of notification to colleagues and the incorporation of patient
specific information, which may be more valuable than our original focus of knowledge-based
resources. Our second phase of design would also address existing workflow issues with the
design and pilot more focused information needs incorporating and mobile tool that has the ability
to capitalize off of existing technology rather than reinventing what already exists. That is, we
would propose that our tool communicate with the HIS and LIS in the hospitals. Limitations to this
study have been a small sample population, the inability to truly test this within a nursing unit in a
real live setting, a lack of observation data and a lack of time for testing the high fidelity prototype.

3
Introduction

Early in the project we had reservations concerning the concept of ChaCha. Three primary
elements of the tool were in question:

• ChaCha lacks authority – there are no citations, no links and no real way to ensure the
answer is correct other than to rely on the administration to choose appropriate subject
experts leaving us to their mercy. User popularity and satisfaction were both cited as a
basis for the viability and authority of this tool in class presentations. However, Scott and I
still object on the basis that self-report measures and user satisfaction can in no way
provide any evidence of the authority of the tool.
• ChaCha provides no method to deter duplicate questions – primarily we saw this as a bad
business model and a failure to capitalize off of previous labor. In short, the company pays
multiple times to provide the same answer. And, sometimes the answer is not even
correct.
• ChaCha lacks scope & focus – given the amount of information in the world, it seems as
though the scope and focus of the tool inevitably will become its fate. We felt a tighter
focus would be more appropriate and manageable.

The above points led us to question whether or not SMS is the best format for this tool. We initially
felt the SMS format would not be the best use for this application and rather a web application
would be more appropriate. However, this still left us with the difficulty of the above problems. We
set forth to develop a web application that could be used on either a mobile device or a desktop
computer which would provide authoritative answers, capitalize from an existing database of
answers and also be significantly tightened in scope.

4
Fieldwork Activity & Data Collected

We began with preliminary surveys and structured telephone interviews involving three teens. This
was an attempt to understand why they used ChaCha and how they used it. We theorized the
largest number of users were teens. However, we knew early on this preliminary data would only
provide us with existing context. The early data showed the design was generally hated, but the
mobility of the product was highly valued. The product was rarely used for anything requiring
accuracy (i.e. homework, health questions, etc.) and was more readily used for trivia, news, and
politics. The teens did not readily have an answer for why they would not use it for homework or
answers where authority would be critical other than to simply suggest they had other resources for
that. Perhaps, the tool was simply not built with the purpose in mind of addressing complex topics.
This would explain why the users were not using it as such. ChaCha is generally a tool for quick
and fast answers to questions that are binary in nature or those questions having a distinct, pointed
answer. Or at least, that is what our data (along with our intuition) suggested.

Our next task was to move to the nursing population and gather data from them. We took two
approaches to our data gathering. We first developed a preliminary survey to gather some data
about their work patterns as well as their access to computers and mobile devices. A 5-question
survey was developed using Survey Monkey and sent to a nursing population numbering 18. The
sample population consisted of primarily nurses who held research-based position in nursing. We
retrieved 8 surveys back for a response rate of 44 percent. The survey questions and results can
be found in Table 1.

Table 1: Literature Search

Question Choices (% selected) N=8

Patient Education (37.5)
1. My most common nursing information Pharmaceutical (25)
needs/questions when working either at bedside Guidelines (62.5)
or in my current position are (choose those that Policy & Procedure (37.5)
apply): Clinical Questions (50)
Pictures (25)

5
 Evidence-Based Research (87.5)
Ask a Colleague (50)
Reference Book (12.5)
Medical Library (50)
2. I use the following tools to answer my Google (75)
questions in nursing: Nursing Consult (50)
CINAHL (62.5)
Medline/PubMed (87.5)
Epocrates (25)
Micromedex (25)
Palm PDA (75)
PC at Nursing Station (87.5)
3. I have access to: Reference Books( 50)
I carry a portable/wireless phone (50)
Medical Library (75)
Nealry 100 percent of the time (37.5)
4. If I could text a question on a mobile device More than half the time (62.5)
and receive an authoritative answer within 5 Less than half the time (0)
minutes, I would use such a service to answer Less than 10 percent of the time (0)
my nursing questions: Never (0)

How quickly I can get the answer (0)

5. I place the most emphasis on the following Having a mobile device that can go with me (0)
value when I have a question in nursing: The authority or correctness of the answer
(87.5)
The source of the answer (12.5)

The results of the survey indicated the importance of authority in the answer rather than how
quickly they can get an answer. The survey also gave us insight into access measures as well as
whether they would use a mobile tool at all. There were some limitations to this survey. We were
not comfortable with a simple self-report and this was also noted in one of our discussion with the
instructor where we were urged to find a method of direct observation. Unfortunately, the time
constraints on this project prevented that from occurring and we were only able to obtain the
survey results along with 3 follow-up interviews. The survey was purposefully short to ensure we
garnered a significant number of responses. However, it did leave some key information out such
as specifics on demographics. The sampling population was small and biased towards nurses who
research which might explain the emphasis on research-based information in question 1 of the
survey. To address these issues, we contacted several of the nurses surveyed and arranged semi-
structured interviews. One in person interview was conducted and two telephone interviews were

6
conducted. We also conducted a literature search to determine the information seeking patterns
and habits of nurses as well as the barriers.

Our interviews allowed us to explore the feasibility of our intended project as well as to explore the
information seeking behavior of nurses in general. We transcribed our notes from the interviews
and developed some broad categories outlining barriers to information seeking in nurses (see
Figure 1).

Figure 1: Barriers to Information Seeking

Our literature review allowed us to identify what motivated nurses to seek information and what
type of information they sought. We searched the literature using both keyword searching and
controlled vocabulary manipulation in both CINAHL and PubMed to identify articles. We identified 9
articles for evaluation and inclusion (see appendix A).1-9 We then tabulated the findings of these
studies to develop an outline of what specifically motivated nurses to search for information and
what type of information they were seeking. The results of this search can be seen in figure 2.

There were some primary barriers we found to seeking information – time, the difficulty of the tool
to use, access to resources and a general lack of experience. Concerning what motivated nurses
to search for information: We determined there were a number of issues involved from personal to

7
professional and organizational the technological. We knew we could not address all barriers and
motivators and thus decided to choose those that would best be answered through the use of
technology. Figure 3 shows the different levels of information needs and how they are best
addressed.

With the qualitative data we had, we were ready to move forward with both goals & requirement
and a conceptual design that would solve some of the issues we had recognized from the
literature.

Figure 2: Motivation to Seek Information

Goals & Requirements

8
As we moved into setting some concrete and practical goals for the project, we realized there was
one final limitation. Figure 3 breaks down information needs of nurses and how we can address
them. Figure 4 illustrates how nurses seek and who they turn to for information. Informatics
approaches are best aligned with those information needs that are binary or questions with a
definitive answer. That being said, we did not know the percentage of information needs. That is, of
the number of questions a nurse might have on any given shift, what percentage would be those
that are quickly answered reference questions and what percentage are questions requiring
research. We define research as searching the medical literature for applicable studies, extracting
those studies and compiling the results to make a decision based on the evidence. This is a
process that can often take days, weeks or months depending on the question.

So we were left to make a decision based on very little to no data. There is virtually nothing in the
literature examining the types of questions nurses have. In retrospect, we would have chosen
clinical decision support as a topic to address since fulfilling those needs are relatively simple and
they are almost always binary questions or possess definitive answers. As it was, we made the
mistake of leaving this open prior to completing the project and giving our presentation. Since the
presentation and the class are designed to be iterative learning processes, we reserve the right to
amend our previous intentions. Our original goals and requirements are as follows:

• Mobile app with crossover capability to web

• Simplicity in design on the frontend
• Simplicity in interface
• Blended information approach with NLM, Automation and Human interaction
• Auto completion tool for speed

We now think and know this is a bit vague. Thus we decided to amend our goals and requirements
for the purpose of this paper.

9
Figure 3: Health Information Levels of Need

Our amended goals and requirements are as follows:

• Develop an application that is both available via web and mobile devices
• Ensure the design and interface offer simplicity
• Implement natural language processing to handle the queries
• Specifically focus on clinical decision support where the reference questions have
definitive answers
• Allow for the capture of more detailed research questions, but provide an indication to the
user that this question will not be answered immediately (this can be provided in the form
of an automated message)
• Use a blended information approach where there is both Natural Language Processing
and human intervention
• Develop a database of questions that can be matched to duplicate incoming questions
10
Figure 4: Information Flow

We also set out some concrete requirements concerning the process and timeline of activities. Our
initial idea was that this would be part of a NLM grant with a number of working parts and persons
in the process. There would be a team of information professionals to handle the human-centered
aspect of the study. However, there would also have to be a certain amount of training up front to
develop information literacy skills in the nurses and help them recognize their own different
information needs. This would enable us to design a system that could supplement the working
parts and early education process in the project.

Salient Scenario 1

Joe is a new nurse on the Cardiovascular Unit (CVU). The CVU is not too quickly paced, but the
patients obviously require more intensive care. Many of them are on ventilators or are recovering
from cardiovascular procedures. One of Joe’s patients – Mr. Smith – will have his chest tube

11
removed today. Joe vaguely remembers chest tube placement and guidelines from school, but has
only done it once on a real person. He needs to know what the procedure for chest tube removal is
as a refresher. He essentially needs to know what steps to take. Joe steps outside of the room and
finds a private area in the nursing station. He pulls a pda out of his pocket and types chest tube
into the application designed to give him mobile answers. Placement is one of the options and Joe
is pleased to find a list of the different steps in the procedure. He looks at them and refreshes his
memory. However, he knows that if he forgets a step he cannot stop in mid-procedure to look at his
pda. Just as he realizes this his nurse manager approaches him explaining one of the more
experienced nurses will be overseeing the procedure with Joe.

Commentary: This is a clear and defined information need. We know there is an answer to the
question and it will not require synthesizing the literature. It is a basic reference question. Joe
needs a refresher and a step-by-step look at the procedure so the attending nurse does not have
to give him a blow-by-blow in front of the patient and scare them. The tool works in the sense that it
is not a direct reference, but a quick overview to supplement what will come. The tool is also not a
replacement for experience as the nurse will have to guide Joe through the procedure and fill the
tacit knowledge gap.

Salient Scenario 2

Sandy is a nurse on an Orthopedic unit where joint replacement and surgeries are common. She
works with a number of different physicians and one common condition they have on the unit is
pain management and how to keep their patients comfortable during the post-surgical period.
Sandy has not been on the unit too long, but she is a relatively experienced nurse. They recently
began joint replacement surgeries and are having trouble with their joint replacement patients
because of the amount of pain they are in. One physician has prescribed medicine on an “as
needed basis” (PRN) rather than around the clock (ATC). One of the physicians Sandy works with
prescribes ATC and Sandy notices those patients seem to do better. She needs to know which is
the best way to prescribe medicines. She types “dosing regimes” into her PDA and gets
information back for specific drugs. She does not know how to formulate this query and after

12
several tries, she uses “pain management” with results that are much too broad and often specific
to the type of medicine. Not having the specific answer she needs, Sandy hits the ask expert
button to work with an expert.

Commentary: This is an ideal scenario, but there are two problems. The first is that the information
she is seeking or rather the question she has is not binary. It is truly a research question for which
there may or may not be a hard and fast answer. This is not the ideal question for this resource.
Unfortunately, there is no way to weed out these types of questions. Second, how many times will
a user encounter a scenario like this before they begin utilizing a workaround and simply hit the
“Ask Expert” button every time? One other point to note is that we must consider the questions in
context. Evidence-based Medicine is not always a single answer issue. There are instances in
which off-label use is appropriate or when we might circumvent the evidence in favor of the nurse
or physician’s expertise and experience. How would we ever address this issue with this tool?

Salient Scenario 3

June is a nurse working on the recovery unit in ICU. Her work is much like Joe’s above where there
are short burst of frenzy and then quietness may follow for an unspecified period of time. There is
ebb and flow to her work. June’s patient was brought in last weekend because they suddenly
experienced a temporary paralysis on the right side of their body. The patient was immediately
given an MRI and the test results indicated a mild stroke. They were given TPA and subsequently
a coagulant. The patient seems to be doing well, but may require some physical therapy. Upon
discharge, the family of the patient has asked for information about strokes, how to manage stroke
victims, how they will recover and what they can do to help. June needs to find something for them
fast since the discharge period is hectic.

Commentary: This is a fairly simple request, but with a lot of different practical issues. Can the
nurse print straight from the mobile device? What sort of information does she give them and what
format would it be on a mobile device. Answering this question would be easy enough, but it is
worthless if you cannot get the information to the patient or their family. Can we link out from this

13
mobile application to Medline or the Mayo Clinic’s patient education resources? There are a lot of
issues to work through here, but the potential exists.

So, we wonder if in the above whether a mobile device could be used. It seems there are a variety
of issues, but this solution has already been proposed in the literature for nursing.1 However, it is
still being prototyped and not been implemented or pilot tested as of our literature search.

Conceptual Design

Our first conceptual design can be seen in Figure 5. Originally we had intended to develop a
blended approach to managing the information needs of the pilot unit. Each question would be
submitted to both local colleagues and to a knowledge database. The database would use natural
language processing to attempt to generate an answer. Any answers generated by either source
would be forwarded to the user. Some sort of method would be necessary to address conflicting
answers by allowing interaction with colleagues. If no answer was generated, the question would
be sent to a live reference expert who would research and forward an answer to the user.

14
Figure 5: First Conceptual Design

The conceptual design itself gave rise to new questions, mainly:

1. What are the exits of the process?

2. How can the colleague requests be minimized appropriately?
3. How can the stress of redundant questions be minimized appropriately?
4. Can conflicting answers be used in an informative way as opposed to a hindrance?

A second iteration of the conceptual design was developed to address these questions and is
shown in Figure 6.

15
Figure 6: Final Conceptual Design

Essentially there would be 3 processes working or 3 levels to the process.

1. We would attempt to match the question with what had already been asked and sat in an
existing knowledge base of questions.
2. If the question had not been previously asked or the answer was not appropriate in the
context, the question would be converted to a query and executed against an existing
knowledge base of raw data. We would propose to achieve this through natural language
processing and return those matching results back to the user.
16
 3. If there was no match or the user indicated they were not satisfied, they would have one of
two options or both. They could ask an expert via a panel of information professionals and
receive an answer that way, or they could ask a local colleague. The history of the
answers received would be forwarded each time a “live” source was queried.

This new design addressed each of our questions directly:

1. What are the exits of the process?

The new design allows the system to be exited at each level and presents the levels in
increasing order of demand placed on system resources. The first level, previously
asked questions, is a simple query answer based on approximate matches from a
local database of questions asked. The second level is an algorithm that searches a
local or regional database for an answer based on a natural language processing of
the user’s question and returns the results of the query. The third level sends the text
question to a “live” source, being either a colleague or a live reference expert using a
research database.

2. How can the colleague requests be minimized appropriately?

Demand on the colleagues would be reduced by giving the user ample opportunities to
query the knowledge bases before asking a colleague.

3. How can the stress of redundant questions be minimized appropriately?

Demand created by redundant questions would need to be discouraged in our original

conceptual design. The modification of presenting previously asked questions without
creating a large stress on the system would instead encourage redundant questioning
if the question was common and at the same time encourage use of the system
without creating much system stress.

17
 4. Can conflicting answers be used in an informative way as opposed to a hindrance?

Forcing the user to query the knowledge bases would not only give a chance to
complete the process without a “live” source but also help focus the user’s question by
presenting related information and starting the dialogue. The dialogue of the user
asking the question and rejecting an answer, perhaps along with a note of explanation,
would be sent when querying a “live” source. This would provide more information to
assist the “live” source with the interpretation of the question, thereby increasing the
expected accuracy of the answers given.

Ideally, whether the nurse asked the colleague or not, we would want to capture all questions to be
reviewed by a unit member at some point in time. This would give the organization the ability to
have some sense as to what needs are being met and which needs could be better addressed
through educational interventions.

Walkthrough Scenario

From the conceptual design, a simple and generic set of use cases was developed alongside a
low-fidelity paper prototype. Three primary use cases were identified:

1. Asking a question
2. Getting a “live” answer
3. Giving an answer

The use cases each illustrate a specific stage of the application of our conceptual design. “Asking
a question” is the use case of a user inputting a question into the interface and walking through
each of the three levels to submit a question to a “live” source. “Getting a ‘live’ answer” is the use
case of a user receiving an answer to the submitted question. “Giving an answer” is then the use
case of a second user receiving the question and submitting an answer.

18
The paper prototype was created as a series of minimalistic, “screenshot” sketches of only the
elements needed to perform each step of the use cases without regard for other functions or
elements that might have to be present in an implemented interface. Alongside each “screenshot,”
a dialogue describing the user’s interaction leading up to and during the use of the interface at that
stage of the use case. Each step of the paper prototype was then directly converted into a mock
interface for both targeted platforms: a standard web browser and a mobile device.

For the standard web browser interface, a high-fidelity prototype was created using simple HTML
for a feasible-looking interface with minimal elements. For the mobile device interface, the Carbide
UI of the Symbian OS was chosen. The Symbian OS is deployed on a large number of both
consumer and enterprise level devices and allows heavy customization and development of
applications. This would provide both the flexibility to meet an individual organization’s process
needs and the customizable security necessary for any system used within a healthcare
organization. The high-fidelity mobile device prototype was created using the elements of the
Carbide UI to create native-resolution screenshots using only the functionality pre-built in the
Carbide UI. In contrast to the standard web browser prototype, the mobile device prototype was
built as an application-mediated process. Instead of using a web service directly, an application
developed specifically for the device would present an interface to the user and would handle the
data transfer between the user and data service in the background. The same elements would be
presented, but the application would be able to run with much lower latency and without having to
convert web content to fit the display.

Visual/Interactive Prototype

Below are the screenshots for the high-fidelity mobile device prototype along with the dialogue
describing the user’s actions. For the paper prototype and the high-fidelity web browser interface
prototype, see the appendix.

19
Use Case 1: Asking a question

1.) User has a question about whether prescribing pain meds "as needed" tends to be "better"
than a Dr-prescribed schedule. User launches UI.

20
2.) User starts to type in question. UI begins parsing and displays possible matches that have
been asked before or seeded into the system.

21
3.) User sees a possible match and taps the question to see the answer.

22
4.) user is not satisfied and continues typing the question. The list of previous questions updates.
user does not see a question that matches and taps the "ASK" button to submit the question.

23
5.) UI parses the question and returns an answer from the local database at the facility/region
level. User is not satisfied with the result and taps the "ASK an expert" button to submit the
question to a "live" expert.

24
6.) UI presents opportunity for user to append a reason why the automated answer was
insufficient. User enters a reason if necessary and taps the "Submit" button.

25
Use Case 2: Getting a "live" answer

1.) UI notifies user of answers received from a "live" source. User taps the "VIEW" button to see
more.

26
2.) UI displays a list of questions asked of a "live" source. User taps the question with an answer
to view the question history.

27
3.) UI displays full history of answers received for the question. user taps "Ask a colleague" to
forward the question to local colleagues and sees "Asking a question (6)". Cycle repeats until user
is satisfied and taps "Question is resolved."

28
Use Case 3: Giving an answer

1.) UI notifies user of questions submitted for answer. User taps the "View"
button to see more.

29
2.) UI displays list of received questions. User taps a question to view the history.

30
3.) UI displays full history of answers received for the question and notes added. User taps
"Submit an answer" button to respond. Alternatively, user could ignore the question and tap the
"Remove from list" button.

31
4.) UI displays the question, and user enters an answer. User taps the "Submit" button to send
the answer.

32
Preliminary User Feedback
Given the time constraints at the end of this project, we did not have an opportunity to garner
formal feedback or truly test our product. We were able to work with a small number of our
colleagues and did send our prototypes to our nursing interviewees. We were able to garner a very
small round of responses and have included an analysis below.

• There does seem to be some confusion as to when they would or should ask a colleague
and when they should ask an expert. Moreover, what is the difference between “expert”
and “colleague” and are there negative connotations associated with using these labels?
• There was general consensus on the aesthetics and simplicity of the interface – overall.
Comments were positive in that respect.
• As we noted above, this tool would be ideal if it were to provide basic information on an
individual patient. This was mentioned by one nurse and several of our colleagues. The
ability to provide vital signs or lab results would be a time saver for healthcare
professionals.
• There were perpetual questions as to how colleagues would manage the alerts if another
member of the unit selected “ask a colleague.” Since we had not thoroughly worked that
out (or at least were rethinking our original idea) we noted this as feedback for
improvement.

Conclusion

We developed and prototyped a design for a mobile device to assist nurses with their health
information needs noting a number of gaps for improvement. We both gathered data and
performed an intensive literature search to identify the information needs of nurses. In the process,
we discovered this idea is being explored elsewhere giving us confidence that with the right
revisions (and a programmer) this project could truly work and is realistic from a conceptual
standpoint.

33
References:

1. Doran DM, Mylopoulos J, Kushniruk A, et al. Evidence in the palm of your hand:
development of an outcomes-focused knowledge translation intervention. Worldviews
Evid Based Nurs. 2007;4(2):69-77.

2. Gilmour JA, Scott SD, Huntington N. Nurses and Internet health information: a
questionnaire survey. J Adv Nurs. 2008;61(1):19-28.

3. Hannes K, Vandersmissen J, De Blaeser L, et al. Barriers to evidence-based nursing:

a focus group study. J Adv Nurs. 2007;60(2):162-171.

4. McCaughan D, Thompson C, Cullum N, Sheldon TA, Thompson DR. Acute care

nurses' perceptions of barriers to using research information in clinical decision-making.
J Adv Nurs. 2002;39(1):46-60.

5. McCleary L, Brown GT. Barriers to paediatric nurses' research utilization. J Adv Nurs.
2003;42(4):364-372.

6. Retsas A. Barriers to using research evidence in nursing practice. J Adv Nurs.

2000;31(3):599-606.

7. Secco ML, Woodgate RL, Hodgson A, et al. A survey study of pediatric nurses' use of
information sources. Comput Inform Nurs. 2006;24(2):105-112.

8. Tannery NH, Wessel CB, Epstein BA, Gadd CS. Hospital nurses' use of knowledge-
based information resources. Nurs Outlook. 2007;55(1):15-19.

9. Xu X, Rocha RA, Bigelow SM, et al. Understanding nurses' information needs and
searching behaviour in acute care settings. AMIA Annu Symp Proc. 2005:839-843.

34
Appendix A
Article
Doran DM, Mylopoulos J, Kushniruk A, et al.
Evidence in the palm of your hand: development of
an outcomes-focused knowledge translation
intervention. Worldviews Evid Based Nurs.
2007;4(2):69-77.
35
Abstract
AIM: The aim of the project was to develop an
electronic information gathering and
dissemination system to support both nursing-
sensitive outcomes data collection and
evidence-based decision-making at the point-
of-patient care. BACKGROUND: With the
current explosion of health-related knowledge,
it is a challenge for nurses to regularly access
information that is most current. The Internet
provides timely access to health information,
however, nurses do not readily use the Internet
to access practice information because of being
task-driven and coping with heavy workloads.
Mobile computing technology addresses this
reality by providing the opportunity for nurses to
access relevant information at the time of
nurse-patient contact. METHOD: A cross-
sectional, mixed-method design was used to
describe nurses' requirements for point-of-care
information collection and utilization. The
sample consisted of 51 nurses from hospital
and home care settings. Data collection
involved work sampling and focus group
interviews. FINDINGS: In the hospital sector,
40% of written information was recorded onto
"personal papers" at point-of-care and later
transcribed into the clinical record. Nurses often
sought information away from the point-of-care;
for example, centrally located health records, or
policy and procedure manuals. In home care,
documentation took place in clients' homes.
The most frequent source of information was
"nurse colleagues." Nurses' top priorities for
information were vital signs data, information on
intravenous (IV) drug compatibility, drug
references, and manuals of policies and
procedures. Implications: A prototype software
system was designed that enables nurses to
use handheld computers to simultaneously
document patients' responses to treatment,
obtain real-time feedback about patient
outcomes, and access electronic resources to
support clinical decision-making. Conclusion:
The prototype software system has the

Gilmour JA, Scott SD, Huntington N. Nurses and
Internet health information: a questionnaire survey. J
Adv Nurs. 2008;61(1):19-28.
36
potential to increase nurses' access to patient
outcomes information and evidence for point-of-
care decision-making.
AIM: This paper is a report of a study to identify
the extent of postgraduate nursing students'
information literacy skills in relation to electronic
media and health information and barriers to
accessing this information. BACKGROUND:
The Internet is a key source of information for a
significant group of patients. However, there is
evidence of quality issues with some Internet
health information sites. Nurses need to be
aware of the range and quality of online health
information so as to assist patients and families
to locate and evaluate relevant and current
information. METHOD: A questionnaire
designed to collect quantitative and qualitative
data was posted to a convenience sample of all
students enrolled in a postgraduate nursing
programme in December 2005. The response
rate was 55.1% or 123 responses. RESULTS:
Most respondents had Internet access at home
and work and believed that access to online
health information resources had improved their
practice. However, some had difficulties in
accessing computers at work and insufficient
time to search for online health information.
Concern was expressed about the quality of
online information, but the majority of
respondents did not assess patient use.
Frequent users of online resources were more
likely to assess patient use. CONCLUSION:
The development of nursing competencies in
accessing and using online resources is a key
precursor to supporting patients and families'
use of the medium. Access to Internet
resources at work, along with training and time
for searching, is necessary for the development
of skills enabling effective use of information
technology.
AIM: This paper reports a study to explore the
barriers to evidence-based nursing among
Flemish (Belgian) nurses. BACKGROUND:
Barriers obstructing the call for an increase in

Hannes K, Vandersmissen J, De Blaeser L, et al.
Barriers to evidence-based nursing: a focus group
study. J Adv Nurs. 2007;60(2):162-171.
McCaughan D, Thompson C, Cullum N, Sheldon
TA, Thompson DR. Acute care nurses' perceptions
of barriers to using research information in clinical
decision-making. J Adv Nurs. 2002;39(1):46-60.
37
evidence-based nursing have been explored in
many countries, mostly through quantitative
study designs. Authors report on lack of time,
resources, evidence, authority, support,
motivation and resistance to change.
Relationships between barriers are seldom
presented. METHODS: We used a grounded
theory approach, and five focus groups were
organized between September 2004 and April
2005 in Belgium. We used purposeful sampling
to recruit 53 nurses working in different
settings. A problem tree was developed to
establish links between codes that emerged
from the data. FINDINGS: The majority of the
barriers were consistent with previous findings.
Flemish (Belgian) nurses added a potential lack
of responsibility in the uptake of evidence-
based nursing, their 'guest' position in a
patient's environment leading to a culture of
adaptation, and a future 'two tier' nursing
practice, which refers to the different education
levels of nurses. The problem tree developed
serves as (1) a basic model for other
researchers who want to explore barriers within
their own healthcare system and (2) a useful
tool for orienting change management
processes. CONCLUSION: Despite the fact
that the problem tree presented is context-
specific for Flanders (Belgium), it gives an
opportunity to develop clear objectives and
targeted strategies for tackling obstacles to
evidence-based nursing.
AIM: To examine the barriers that nurses feel
prevent them from using research in the
decisions they make. BACKGROUND: A
sizeable research literature focusing on
research utilization in nursing has developed
over the past 20 years. However, this literature
is characterized by a number of weaknesses:
self-reported utilization behaviour; poor
response rates and small, nonrandom sampling
strategies. DESIGN: Cross-case analysis
involving anonymised qualitative interviews,
observation, documentary audit and Q
methodological modelling of shared
subjectivities amongst nurses. The case sites
were three large acute hospitals in the north of
 England. One hundred and eight nurses were
interviewed, 61 of whom were also observed for
a total of 180 h, and 122 nurses were involved
in the Q modelling exercise (response rate of
64%). RESULTS: Four perspectives were
isolated that encompassed the characteristics
associated with barriers to research use. These
related to the individual, organization, nature of
research information itself and environment.
Nurses clustered around four main
perspectives on the barriers to research use:
(1) Problems in interpreting and using research
products, which were seen as too complex,
'academic' and overly statistical; (2) Nurses
who felt confident with research-based
information perceived a lack of organizational
support as a significant block; (3) Many nurses
felt that researchers and research products lack
clinical credibility and that they fail to offer the
desired level of clinical direction; (4) Some
nurses lacked the skills and, to a lesser degree,
the motivation to use research themselves.
These individuals liked research messages
passed on to them by a third party and sought
to foster others' involvement in research-based
practice, rather than becoming directly involved
themselves. CONCLUSIONS: Rejection of
research knowledge is not a barrier to its
application. Rather, the presentation and
management of research knowledge in the
workplace represent significant challenges for
clinicians, policy-makers and the research
community.

BACKGROUND: A number of studies have

McCleary L, Brown GT. Barriers to paediatric nurses' study this. Reported barriers have been
research utilization. J Adv Nurs. 2003;42(4):364-
372.
investigated barriers to research utilization
among nurses in various countries, and
standardized scales have been validated to
categorized as individual, organizational and
environmental, with organizational
characteristics generally accounting for more
variance. However, information about research
utilization among paediatric nurses is lacking.
AIM: The objective of the research reported
here was to investigate barriers to research
utilization and relationships between those
barriers and participation in research, self-

38

Retsas A. Barriers to using research evidence in
nursing practice. J Adv Nurs. 2000;31(3):599-606.
39
reported research utilization and education
among paediatric nurses. DESIGN: A survey of
all nurses in a paediatric teaching hospital; 176
nurses (33.3%) responded. Two standardized
measures were used, the Barriers Scale and
the Edmonton Research Orientation Scale.
RESULTS: Lack of time to read research was
the most frequently cited barrier to using
research and administrators not allowing
implementation was the least frequently cited.
Characteristics of the communication and of the
setting were more likely to be cited as barriers
to research use than were characteristics of the
nurse. Nurses who reported higher levels of
actual research use were slightly less likely to
see characteristics within themselves as
barriers. Those who had taken a course about
reading or using research were more likely to
see the organization as a barrier. Barriers to
research use were not associated with self-
reported understanding of research.
CONCLUSIONS: These results are congruent
with previous findings that implementing
research in practice is a complex process. They
indicate that individual nurses' knowledge about
research may not be as important as the
process by which organizations implement
research. However, the Barriers Scale
measures general perceptions about barriers to
research utilization and not nurses' specific
experiences with barriers to implementing
particular research.
Barriers to using research evidence in nursing
practice The nursing profession has long
recognized the importance of research as an
essential basis for its development. More
recently, the movement supporting evidence-
based practice has brought this point into focus.
However, relatively little research has been
conducted to identify factors that interfere with
the ability of nurses to base their practice on
research evidence. Using factor analysis
procedures, barriers to the use of research
evidence in practice which were perceived to
be present by 400 registered nurses working in

Secco ML, Woodgate RL, Hodgson A, et al. A
survey study of pediatric nurses' use of information
sources. Comput Inform Nurs. 2006;24(2):105-112.
Tannery NH, Wessel CB, Epstein BA, Gadd CS.
Hospital nurses' use of knowledge-based
information resources. Nurs Outlook. 2007;55(1):15-
19.
40
an Australian hospital, were grouped under four
main factors. These were accessibility of
research findings, anticipated outcomes of
using research, organizational support to use
research and support from others to use
research. The most important factor was
perceived to be organizational support,
particularly in relation to providing time to use
and conduct research.
This survey study explored use of different
information sources among a convenience
sample of 113 bedside pediatric nurses. The
study was guided by three interrelated
concepts: types of information sources, levels
of evidence, and computer skill. The Nursing
Information Use Survey measured use of
information sources, impact of information
sources on nursing care, barriers to
information, and expectations that a
computerized clinical desktop or patient
information management system would
improve patient care. Significant correlations
between use of interpersonal and non-
computer-based information and non-computer-
and computer-based information supported the
conceptual model. Use of traditional, non-
computer information sources such as
textbooks and print-based journals was higher
among baccalaureate, compared with diploma,
prepared nurses. Nurses with greater computer
and online searching skill used more computer-
based information. Findings suggested that
strategies to improve nurses' computer and
information searching skills may promote use of
higher-level evidence in planning nursing care.
The purpose of this study was to evaluate the
information-seeking practices of nurses before
and after access to a library's electronic
collection of information resources. This is a
pre/post intervention study of nurses at a rural
community hospital. The hospital contracted
with an academic health sciences library for
access to a collection of online knowledge-
based resources. Self-report surveys were

Xu X, Rocha RA, Bigelow SM, et al. Understanding
nurses' information needs and searching behaviour
in acute care settings. AMIA Annu Symp Proc.
2005:839-843.
41
used to obtain information about nurses'
computer use and how they locate and access
information to answer questions related to their
patient care activities. In 2001, self-report
surveys were sent to the hospital's 573 nurses
during implementation of access to online
resources with a post-implementation survey
sent 1 year later. At the initiation of access to
the library's electronic resources, nurses turned
to colleagues and print textbooks or journals to
satisfy their information needs. After 1 year of
access, 20% of the nurses had begun to use
the library's electronic resources. The study
outcome suggests ready access to knowledge-
based electronic information resources can
lead to changes in behavior among some
nurses.
We report the results of a pilot study designed
to describe nurses' information needs and
searching behaviour in acute care settings.
Several studies have indicated that nurses
have unmet information needs while delivering
care to patients. AIM: Identify the information
needs of nurses in acute care settings.
METHODS: Nurses at three hospitals were
asked to use an information retrieval tool (CPG
Viewer). A detailed log of their interactions with
the tool was generated. RESULTS AND
CONCLUSIONS: Our findings suggest that
nurses' information needs are different from
what is reported in the literature in terms of
physicians' information needs. Questions
regarding a nursing procedure or protocol were
the most common needs nurses had.
Appendix B
High-Fidelity Prototype - Standard Web Browser Interface
Asking a Question (1)
High-Fidelity Prototype - Standard Web Browser Interface
Asking a Question (2)
High-Fidelity Prototype - Standard Web Browser Interface
Asking a Question (3)
High-Fidelity Prototype - Standard Web Browser Interface
Asking a Question (4)
High-Fidelity Prototype - Standard Web Browser Interface
Asking a Question (5)
High-Fidelity Prototype - Standard Web Browser Interface
Asking a Question (6)
High-Fidelity Prototype - Standard Web Browser Interface
Receiving an answer (1)
High-Fidelity Prototype - Standard Web Browser Interface
Receiving an answer (2)
High-Fidelity Prototype - Standard Web Browser Interface
Receiving an answer (3)
High-Fidelity Prototype - Standard Web Browser Interface
Giving an answer (1)
High-Fidelity Prototype - Standard Web Browser Interface
Giving an answer (2)
High-Fidelity Prototype - Standard Web Browser Interface
Giving an answer (3)
High-Fidelity Prototype - Standard Web Browser Interface
Giving an answer (4)