Brilliant Baby Brainiacs (BBB) - Pediatric BrainTumors: Assessing Healthcare Provider Knowledge
Item Type text; Electronic Dissertation
Authors Tong, Amanda Kai-Lai
Publisher The University of Arizona.
Rights Copyright © is held by the author. Digital access to this materialis made possible by the University Libraries, University of Arizona.Further transmission, reproduction or presentation (such aspublic display or performance) of protected items is prohibitedexcept with permission of the author.
Download date 17/06/2021 07:18:38
Link to Item http://hdl.handle.net/10150/593599
http://hdl.handle.net/10150/593599
BRILLIANT BABY BRAINIACS (BBB) – PEDIATRIC BRAIN TUMORS:
ASSESSING HEALTHCARE PROVIDER KNOWLEDGE
by
Amanda Kai-Lai Tong
________________________"Copyright © Amanda Kai-Lai Tong 2015
A DNP Project Submitted to the Faculty of the
COLLEGE OF NURSING
In Partial Fulfillment of the Requirements For the Degree of
DOCTOR OF NURSING PRACTICE
In the Graduate College
THE UNIVERSITY OF ARIZONA
2 0 1 5
2
THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE
As members of the DNP Project Committee, we certify that we have read the DNP
Project prepared by Amanda Kai-Lai Tong entitled Brilliant Baby Brainiacs (BBB) –
Pediatric Brain Tumors: Assessing Healthcare Provider Knowledge and recommend that
it be accepted as fulfilling the DNP Project requirement for the Degree of Doctor of
Nursing Practice.
_____________________________________________Date: October 14, 2015 Gloanna Peek, PhD, RN, CPNP _____________________________________________Date: October 14, 2015 Ida (Ki) Moore, DNSc, RN, FAAN _____________________________________________Date: October 14, 2015 Cindy Rishel, PhD, RN, OCN Final approval and acceptance of this DNP Project is contingent upon the candidate’s submission of the final copies of the DNP Project to the Graduate College. I hereby certify that I have read this DNP Project prepared under my direction and recommend that it be accepted as fulfilling the DNP Project requirement. _____________________________________________Date: October 14, 2015 DNP Project Director: Gloanna Peek, PhD, RN, CPNP
3
STATEMENT BY THE AUTHOR
This DNP Project has been submitted in partial fulfillment of requirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.
Brief quotations from this DNP Project are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the copyright holder.
SIGNED: Amanda Kai-Lai Tong
4
ACKNOWLEDGEMENTS
First and foremost, I would like to thank my advisor and committee chair Dr. Gloanna Peek for her support and guidance throughout my graduate degree career. I would also like to thank my committee members Dr. Ki Moore and Dr. Cindy Rishel. All three of these professors at the College of Nursing provided expert suggestions on my DNP Project that benefitted all the children I care for as a registered nurse and future pediatric nurse practitioner. I would also like to express my appreciation and gratitude to all of my co-workers at Phoenix Children’s Hospital that continuously supported me throughout my nursing career and my academic career. They were supportive with their words of wisdom, friendship, laughter, and their love of caring for children and their families. I am honored to have taken care of a lot of children and families throughout my nursing career. A few of the patients and families that I cared for inspired this DNP Project. I will never forget the valuable life lessons they all taught me as I cared for them. I would also like to mention that my strong faith and trust in God allowed me to complete this rigorous program and project. Finally, I would like to thank my family and friends, especially my mother, father, and brother. Their love, support, encouragement, and guidance allowed me to complete this DNP Project and my doctoral degree.
5
DEDICATION
To all children and families that have ever had to experience the diagnosis of a
brain tumor and to all the nurses that have cared for these children and families. Most
importantly, to the 8-year old girl and 5-month old baby boy and their lovely families that
inspired me to research about pediatric brain tumors.
6
TABLE OF CONTENTS
LIST OF FIGURES .............................................................................................................9
LIST OF TABLES .............................................................................................................10
ABSTRACT .......................................................................................................................11
CHAPTER I. BACKGROUND AND SIGNIFICANCE ..............................................12
Introduction ......................................................................................................................12 Clinical Problem ...............................................................................................................13
Overview ...............................................................................................................13 Common Signs and Symptoms of Pediatric Brain Tumors .............................13 Brain Tumor Histology and Brain Tumor Location ........................................14 Pre-Diagnostic Symptom Interval (PSI) ............................................................15 Parental Perceptions ............................................................................................16 Summary ...............................................................................................................16
Clinical Guideline to Assist Healthcare Providers ........................................................17 Head Smart Campaign from the United Kingdom ...........................................17 Knowledge Gap ....................................................................................................18
Significance to Advanced Practice Nursing ...................................................................19 Purpose and Aims of DNP Project .................................................................................20
CHAPTER II. REVIEW OF LITERATURE ...............................................................21
Pediatric Brain Tumor Clinical Presentation ...............................................................21 Common Signs and Symptoms ...........................................................................21 Common Misdiagnoses ........................................................................................25 Tumor Histology and Tumor Location Association with Diagnostic Delay and Signs and Symptoms ....................................................................................25
Diagnostic Delays and Pre-Diagnostic Symptom Interval (PSI) .................................29 Pre-Diagnostic Symptom Interval (PSI) ............................................................29 Parental Delay ......................................................................................................30 Doctor’s Delay ......................................................................................................31 Summary of Diagnostic Delays and Pre-Diagnostic Symptom Interval (PSI)31 Primary Care Providers Versus Specialist Doctors ..........................................34 Number of Consultations or Visits to Healthcare Providers Prior to Diagnosis ...............................................................................................................35 Diagnostic Delay of Pediatric Brain Tumors in Comparison to Other Childhood Cancers ..............................................................................................36
Cost-effectiveness of Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) to Diagnose Pediatric Brain Tumors ...................................................37 Does Prolonged Pre-Diagnostic Symptom Interval (PSI) Decrease Survival? ...........39 Parental Perception and Experiences ............................................................................41
7
TABLE OF CONTENTS - Continued
Healthcare Provider Guideline to Assist in Diagnosis of Pediatric Brain Tumors ....42 Synthesis of Literature ....................................................................................................45
CHAPTER III. METHODOLOGY ...............................................................................48
Study Aims ........................................................................................................................48 Model to Implement Guideline: ACE Star Model of Knowledge Transformation ...48
Star Point 1: Discovery Research ......................................................................50 Star Point 2: Evidence Summary ......................................................................50 Star Point 3: Translation to Guidelines ............................................................50 Star Point 4: Practice Integration .....................................................................50 Star Point 5: Process, Outcome Evaluation .....................................................51
Applicability of ACE Star Model of Knowledge Transformation to the Brain Pathways Guideline .........................................................................................................51
Star Points 1-3: The Brain Pathways Guideline .................................................51 Star Point 4 and 5 .................................................................................................52
Methods .............................................................................................................................52 Summary of Expert Evaluations of Pre-Test and Post-Test ........................................55 Sample ...............................................................................................................................57 Protection of Human Subjects and Ethical Considerations .........................................58 Data Collection .................................................................................................................59 Data Analysis ....................................................................................................................59
CHAPTER IV. RESULTS ..............................................................................................60
Description of Sample ......................................................................................................60 Scores on Pre-Test and Post-Test ...................................................................................61 Comparison of Primary Care PNPs to Acute Care PNPs ............................................61 Post-Test Only Questions ................................................................................................62 Wilcoxon Signed Rank Test and Null Hypothesis ........................................................63
CHAPTER V. DISCUSSION AND IMPLICATIONS .................................................64
DNP Project Goals ...........................................................................................................64 Short-Term Goal: Increase Knowledge of Common Signs and Symptoms of Pediatric Brain Tumors .......................................................................................64 Long-Term Goal: Decrease Pre-Diagnostic Symptom Interval (PSI) ............65
Integration of Implementation Model ............................................................................66 Evaluation Questionnaire by Participants ........................................................66
Strengths and Limitations ...............................................................................................67 Relevance to Advanced Practice Nursing ......................................................................69 Future Research and Improvements ..............................................................................70 Conclusion ........................................................................................................................72
8
TABLE OF CONTENTS - Continued
APPENDIX A ....................................................................................................................73
APPENDIX B ....................................................................................................................75
APPENDIX C ....................................................................................................................77
APPENDIX D ....................................................................................................................81
APPENDIX E ....................................................................................................................85
APPENDIX F ....................................................................................................................88
APPENDIX G ....................................................................................................................91
APPENDIX H ....................................................................................................................94
APPENDIX I .....................................................................................................................98
APPENDIX J ...................................................................................................................100
REFERENCES ................................................................................................................102
9
LIST OF FIGURES
FIGURE 1. The Brain Pathways Guideline: A Guideline To Assist Healthcare
Professionals In The Assessment Of Children Who May Have A Brain Tumour (The
Brain Pathways Guideline) ................................................................................................74
FIGURE 2. Decision Tree for Children With Headache ...................................................39
FIGURE 3. ACE Star Model of Knowledge Transformation ............................................49
FIGURE 4. Head Smart Symptom Card ............................................................................76
10
LIST OF TABLES
TABLE 1. Summary of Pre-Diagnostic Symptom Interval (PSI), Parental Delay, and
Doctor’s Delay From Multiple Research Studies ..................................................................33
TABLE 2. Overall Survival Probabilities (OS) and Progression Free Survival
Probabilities (PFS) in Pediatric Brain Tumor Patients ..........................................................41
TABLE 3. Expert Evaluation by Pediatric Intensive Care Unit Attending with Expertise
in Neuro-Critical Care ............................................................................................................55
TABLE 4. Expert Evaluation by Pediatric Neurosurgeon .....................................................56
TABLE 5. Expert Evaluation by Pediatric Nurse Practitioner in Neurosurgery Specialty ...57
TABLE 6. Descriptive Information of Sample ......................................................................60
TABLE 7. Pre-Test and Post-Test Average Scores ...............................................................61
TABLE 8. Comparison of Primary Care PNPs and Acute Care PNPs ..................................62
11
ABSTRACT
Background: Brain tumors are the most common solid tumors found in children. Current
research is determining whether diagnosing brain tumors earlier will help improve
prognosis and reduce long-term deficits; however, childhood brain tumors are often
diagnosed late with a median time of 1-4 months from onset of symptoms. Prolonged
symptom intervals before diagnosis have been associated with life-threatening risks,
neuro-cognitive disabilities, and detrimental professional relationships between
healthcare providers and families. Pediatric brain tumor clinical presentations are often
non-specific and resemble less serious illnesses; therefore, healthcare providers are
failing to include this in their differential diagnoses list.
Purpose: To assess healthcare provider knowledge of signs and symptoms of pediatric
brain tumors using The Brain Pathways Guideline.
Methods: A one group pre-test and post-test e-mailed separately to nurse practitioners
that have active membership in National Association of Pediatric Nurse Practitioners
(NAPNAP) Arizona Chapter.
Results: The Wilcoxon Signed Rank Test revealed that the matched test scores were not
statistically significant (p=0.157) after viewing The Brain Pathways Guideline
educational materials.
Conclusion: The results of this study did not show a statistically significant difference in
the test scores and therefore it cannot be concluded that presenting an evidence-based
guideline to assist healthcare providers to assess and diagnose patients with brain tumors
will be helpful to improve pre-diagnostic symptom intervals.
12
CHAPTER I: BACKGROUND AND SIGNIFICANCE
Introduction
Brain tumors are the most common solid tumors found in children (Dobrovoljac,
Hengartner, Boltshauser, & Grotzer, 2002; Klitbo, Nielson, Illum, Wehner, & Carlsen,
2011; Wilne et al., 2012; Wilne et al., 2007; Wilne et al., 2010). About 25% of all
cancers are brain tumors in children ages 1-19 years old and 60% of survivors are left
with devastating disabilities (Dobrovoljac et al., 2002; Klitbo et al., 2011; Wilne et al.,
2012; Wilne et al., 2007; Wilne et al., 2010). Current research is determining whether
diagnosing brain tumors earlier will help improve the prognosis and reduce long-term
deficits; however, childhood brain tumors are most often diagnosed late with a median
time of one to four months from the onset of symptoms (Dobrovoljac et al., 2002;
Dörner, Fritsch, Stark, & Mehdorn, 2007; Edgeworth, Bullock, Bailey, Gallagher, &
Crouchman, 1996; Flores, Williams, Bell, O’Brien, & Ragab, 1986; Klitbo et al., 2011;
Pollock, Brischer, & Vietti, 1991; Reulecke, Erker, Fieldler, Niederstadt, & Kurelmann,
2008; Thulesius, Pola, & Håkansson, 2000; Wilne et al., 2012; Wilne et al., 2007; Wilne
et al., 2010). Prolonged symptom intervals before childhood brain tumor diagnosis in
pediatric patients have been associated with increased life-threatening risks, neurological
disability complications, and worsening cognitive abilities in survivors (Wilne et al.,
2012). Detrimental consequences of late diagnosis occur in the professional relationship
domain between healthcare providers and families due to the family’s perception that the
medical responses are incompetent and delayed (Dixon-Woods, Findlay, Young, Cox, &
Heney, 2001; Wilne et al., 2012).
13
Clinical Problem
Overview
The delay in diagnosis of childhood brain tumors dates back to the pre-computed
tomography (CT) and pre-magnetic resonance imaging (MRI) technologies (Edgeworth
et al., 1996). Despite the availability and advances in neuroimaging technologies over
past decades, timely diagnosis of children suspected of having a brain tumor still remains
problematic (Dobrovoljac et al., 2002; Edgeworth et al., 1996). The initial signs and
symptoms of brain tumors are non-specific and often are similar to other common and
less serious illnesses such as gastroenteritis, migraines, or behavioral problems (Wilne et
al., 2007; Wilne et al., 2006). The main clinical problem is that healthcare providers have
difficulty identifying the signs and symptoms associated with pediatric brain tumors due
to the non-specific presentations and failure to consider it as one of the differential
diagnoses (Wilne et al., 2007; Wilne et al., 2010). Consequences of a delayed diagnosis
are that children continue to be symptomatic for a few months and sometimes these
symptoms progress to life-threatening emergencies (Wilne et al., 2012).
Common Signs and Symptoms of Pediatric Brain Tumors
The presentation of pediatric brain tumors is often non-specific with symptoms
mimicking less serious illnesses (Wilne et al., 2010). Wilne et al. (2010) completed a
systematic review and meta-analysis of common signs and symptoms of pediatric brain
tumors. Of the 74 studies that met inclusion criteria, 61 of those studies described signs
and symptoms at diagnosis of 4,171 children with intracranial tumors. The most
common signs and symptoms by listing in order of decreasing frequency are: headache
14
(32%), nausea and vomiting (32%), abnormal gait or coordination (27%), papilloedema
(13%), seizures (13%), unspecified signs and symptoms of increased intracranial pressure
(10%), squint (7%), changes in behavior of school performance (7%), macrocephaly
(7%), unspecified cranial nerve palsies (7%), lethargy (6%), abnormal eye movements
(6%), hemiplegia (6%), weight loss (5%), focal motor weaknesses (5%), unspecified
visual or eye abnormalities (5%), and altered level of consciousness (5%). The most
frequent signs and symptoms that healthcare providers may encounter will be headache,
nausea and vomiting, or abnormal gait or coordination. These symptoms alone may
mislead healthcare providers to consider less serious diagnoses on their differential
diagnoses list. Reulecke et al. (2008) found that healthcare providers are aware that a
delay in the diagnosis of a brain tumor may lead to further progression of the tumor thus
worsening the outcome and posing additional risk of brain damage due to prolonged
increased intracranial pressure (ICP).
Brain Tumor Histology and Brain Tumor Location
Other factors that affect the presentation of a child suspected of having a brain
tumor are tumor histology, tumor location, and especially the age of the child (Pollack,
1999). Childhood brain tumors are classified as supratentorial or infratentorial.
Supratentorial tumors develop above the tentorium cerebelli and infratentorial tumors
develop below the tentorium cerebelli (Pollack, 1999). About 60% of children develop
brain tumors in the infratentorial region and supratentorial tumors are predominant in
infants (Pollack, 1999). Supratentorial tumor presentations reflect the compression and
infiltration of adjacent structures; therefore, the patient may have signs and symptoms of
15
increased ICP (Pollack, 1999). Infratentorial tumors have a variety of presentations
depending on the tumor type, but these tumors commonly occlude the fourth ventricle
leading to obstructive hydrocephalus and increased ICP signs and symptoms (Pollack,
1999). Depending on the histology of the tumor, signs and symptoms may vary. Signs
and symptoms of benign tumors progress insidiously; conversely, signs and symptoms
for aggressive tumors manifest more rapidly (Pollack, 1999; Reulecke et al., 2008). For
these reasons, healthcare providers need to be cognizant of the common signs and
symptoms that might lead them to a differential diagnosis of a pediatric brain tumor.
Pre-Diagnostic Symptom Interval (PSI)
The period that precedes uncertainty of a brain tumor diagnosis has been termed
pre-diagnostic symptom interval (PSI), which is defined as the time from initial onset of
signs and symptoms to official diagnosis of a brain tumor by neuroimaging or biopsy
(Dobrovoljac et al., 2002; Dörner et al., 2007; Edgeworth et al., 1996; Flores et al., 1986;
Klitbo et al., 2011; Pollock et al., 1991; Wilne et al., 2012; Wilne et al., 2010). The PSI
can be divided into two stages of delay. One stage is parental delay, which is defined as
the interval from the onset of the first sign and symptom to first visit with a healthcare
provider (Dobrovoljac et al., 2002). The second stage is doctor’s delay, which is defined
as the interval from the first consultation to the diagnosis by neuroimaging technology
(CT or MRI) (Dobrovoljac et al., 2002). The delay in diagnosing pediatric brain tumors
ensues in both stages and regardless of when the delay is occurring, childhood brain
tumors are still diagnosed late (Dobrovoljac et al., 2002).
16
Parental Perceptions
Parents have expressed the interval from noticing symptoms to finally receiving
an official diagnosis for their child as the most distressing experience (Dixon-Woods et
al., 2001). Many parents experience disputes with their healthcare providers initially.
Dixon-Woods et al. (2001) define disputes as “disagreements between parents and
doctors about the seriousness of [their child’s] symptoms and the need for further
investigations” (p. 670). These parents may have to argue with their providers to demand
more tests to be done because of their rejection of the provider’s common sense
diagnosis. The experience of finally arriving at the diagnosis of a brain tumor after many
disputes with healthcare providers offers relief because at least there is an identifiable
cause for the unusual signs and symptoms. For the parents that do not experience many
disputes with healthcare providers, they are shocked and stunned by the child’s tumor
diagnosis whether it is benign or malignant (Dixon-Woods et al., 2001). Parents are
advocates for their children and healthcare providers are also advocates for their patients.
Lengthy disputes with healthcare providers leave parents with the perception that
healthcare providers are incompetent due to their inadequate and inappropriate responses
to their child’s signs and symptoms (Dixon-Woods et al., 2001).
Summary
Advances in technologies and the education of healthcare providers have not
resulted in earlier diagnosis of pediatric brain tumors as evidenced by a continued lengthy
PSI. The non-specific presentation of children suspected of having a brain tumor to
healthcare providers continues to pose a problem because it puts children at risk for
17
further neurological damage and deficits. Negative parental perception of healthcare
providers and the healthcare system may be propagated when parents need to advocate
for their child through disputes. The aforementioned issues with delayed diagnosis
serves as this DNP Project’s significance to implement a guideline to assist healthcare
providers and decrease the PSI as the main outcome measure in primary care settings
(doctor’s offices, urgent care centers, clinics, and emergency departments). The main
clinical problem this DNP Project addresses is that healthcare providers are not
recognizing the signs and symptoms associated with pediatric brain tumors and are not
considering this as a diagnosis on the differential diagnosis list. The purpose of this DNP
Project is to expose healthcare providers to an evidence-based clinical practice guideline
to assess knowledge of pediatric brain tumors to achieve a long-term goal of decreasing
the PSI.
Clinical Guideline to Assist Healthcare Providers
Head Smart Campaign from the United Kingdom
In the United Kingdom, collaboration between healthcare professionals and
families were formed due to their common concern about the prolonged time it was
taking for children with brain tumors to be diagnosed. The response to this concern was
the launch of a national campaign called the Head Smart Campaign (Head Smart, 2011)
after The Children’s Brain Tumour Research Centre at Nottingham University (Walker et
al., 2008) developed a specific evidence-based guideline to aid practitioners in
identifying and assessing children suspected of having a brain tumor. The Head Smart
Campaign has specific aims to enhance the awareness of signs and symptoms of pediatric
18
brain tumors and reduce the delay of diagnosis. The goal of this campaign is to reduce
the pre-diagnostic symptom interval to six weeks, when research has shown that
diagnosis of brain tumors have been taking longer than three months (Head Smart, 2011).
The campaign aims to decrease the PSI at all stages, which include: delay in seeing a
healthcare provider, delay in healthcare providers’ recognition of the signs and
symptoms, and delay in referral to appropriate secondary care such as neuroimaging
appointments (CT or MRI), neurosurgery specialty providers, or neuro-oncology
specialty providers (Head Smart, 2011). Currently, the United States does not have the
same national initiative, goal, and campaign compared to the United Kingdom, though
the delay in diagnosis of pediatric brain tumors exists worldwide in developed countries.
Knowledge Gap
Pediatric brain tumors are the most common tumors discovered in children;
however, during a healthcare providers’ career, he/she may only see one case
(Dobrovoljac et al., 2002; Wilne et al., 2012). Without a current implemented guideline
to assist healthcare providers in assessing children that may have a brain tumor, diagnosis
may continue to be delayed. Walker et al. (2008) and Wilne et al. (2010) conducted
multiple research studies to identify the most common signs and symptoms of pediatric
brain tumors and a multi-disciplinary and Delphi process to develop a guideline called
The Brain Pathways Guideline: A Guideline To Assist Healthcare Professionals In The
Assessment Of Children Who May Have A Brain Tumour (The Brain Pathways
Guideline), which will aid healthcare providers in diagnosing pediatric brain tumors (See
Appendix A Figure 1). This guideline assists healthcare providers to consider a brain
19
tumor diagnosis when a child presents with headache, nausea and/or vomiting, visual
signs and symptoms, motor signs and symptoms, growth and development abnormalities,
behavioral changes, diabetes insipidus, seizures, or altered consciousness. It also
recommends Central Nervous System (CNS) imaging depending on the clinical
presentation and a focused history and physical examination of the neurological system
(past medical history, predisposing factors, history of present illness, visual system,
motor system, height and weight, head circumference for less than 2 years old,
psychomotor development for less than 5 years old, and pubertal status) (Walker et al.,
2008; Wilne et al., 2010). To date, there is no evidence that this guideline has been
implemented into healthcare settings and therefore there are no available results of its
effectiveness to decrease the PSI (Walker et al., 2008; Wilne et al., 2010). The gap in
knowledge among healthcare providers regarding the diagnosis of brain tumors in
children continues to be a concern due to the lack of consideration of a brain tumor as
one of the differential diagnosis at the initial patient visit. This further delays referral for
neuroimaging for diagnosis and prolongs the PSI that may progress to worsening of
symptoms and distressing experiences for patients and families.
Significance to Advanced Practice Nursing
The Advanced Practice Nurse (APN) especially pediatric specialty nurse
practitioner is responsible for providing care throughout a child’s life. APNs are
responsible for recognizing the signs and symptoms of common and uncommon
childhood diagnoses. Being cognizant of brain tumor signs and symptoms with the
assistance of a clinical guideline may help reduce the PSI and allow children with
20
possible brain tumors to receive appropriate care earlier. Providing the APN with a
guideline for assessing pediatric patients for a brain tumor can dramatically affect the
patient and family’s perception of nurse practitioners and the healthcare system.
Implementing the existing clinical guideline available from the Head Smart Campaign
will generate data needed to evaluate whether or not this guideline will decrease the pre-
diagnostic symptom interval in diagnosing children with a suspected brain tumor.
Purpose and Aims of DNP Project
This DNP Project describes an implementation project to assess healthcare
provider knowledge of pediatric brain tumor patients with a long-term goal and aims of
reducing the pre-diagnostic symptom interval by using The Brain Pathways Guideline: A
Guideline To Assist Healthcare Professionals In The Assessment Of Children Who May
Have A Brain Tumour (The Brain Pathways Guideline) developed by The Children’s
Brain Tumour Research Centre at Nottingham University to assist healthcare providers in
assessing and diagnosing children who may have a brain tumor (Walker et al., 2008;
Wilne et al., 2010). A specific Population, Intervention, Comparison, Outcome (PICO)
question has been developed to guide literature search and study design: In children
suspected of having a brain tumor (P), does implementing a diagnostic guideline to assist
healthcare providers (I), compared to not using a guideline (C), increase knowledge about
pediatric brain tumor clinical presentation (short-term) and decrease the pre-diagnostic
symptom interval (long-term) (O)? This DNP Project describes an implementation
project of online educational materials targeting pediatric nurse practitioners with a pre-
test and post-test to assess the short-term goal of knowledge gain.
21
CHAPTER II: REVIEW OF LITERATURE
Pediatric Brain Tumor Clinical Presentation
Common Signs and Symptoms
Numerous retrospective studies have sought to identify the most common signs
and symptoms children with brain tumors have presented with to healthcare providers
(Dörner et al., 2007; Hyashi et al., 2010; Klitbo et al., 2011; Mehta, Chapman, McNeely,
Walling, & Howes, 2002; Reulecke et al., 2008; Wilne et al., 2012; Wilne et al., 2007;
Wilne et al., 2006). Researchers have categorized the most common signs and symptoms
in correlation with tumor histology, tumor location, increased ICP, number of symptoms
initially to number of symptoms at the time of diagnosis, and progression of signs and
symptoms. The consensus among these retrospective studies is that the most common
signs and symptoms are non-specific such as headache, nausea and vomiting, and
abnormal gait or coordination (Dörner et al., 2007; Hyashi et al., 2010; Klitbo et al.,
2011; Mehta et al., 2002; Reulecke et al., 2008; Wilne et al., 2012; Wilne et al., 2007;
Wilne et al., 2006).
Mehta et al. (2002) conducted a retrospective and prospective study in Eastern
Canada of 104 pediatric brain tumor symptoms by analyzing information from medical
charts and interviews from families. Sixty-six percent of children experienced nausea
and vomiting with nine patients that were less than 4-years old, 63% of children
complained of headaches, and 50% of children had some change in their behavior noticed
by the parents.
22
Wilne et al. (2006) completed a review of 200 case note reviews of patients that
presented with a CNS tumor in the United Kingdom. The most common first symptom
among these patients was headache (41%) with most of these headaches either occurring
early in the morning or nocturnally and the second most common symptom was vomiting
(12%). Special consideration that these researchers discovered was that 21% of the
patients that were 3-years old or younger were less likely to present with headaches or
seizures, but more commonly to present with behavioral problems compared to older
children.
Dörner et al. (2007) retrospectively analyzed 50 pediatric patients with posterior
fossa tumors in Germany over a 4-year period and found that the most common
symptoms healthcare providers documented during chart review were headache, nausea
and vomiting, ataxia, cranial nerve deficits, and hydrocephalus.
Reuleck et al. (2008) retrospectively analyzed records of 245 pediatric brain
tumor patients over a 24-year period in Germany. Comparable to previous mentioned
studies, the most common signs and symptoms included headaches and vomiting
initially.
Hyashi et al. (2010) analyzed medical records of 60 pediatric brain tumor patients
in Japan retrospectively for initial signs and symptoms and the progression and number
of the signs and symptoms. The most common initial signs and symptoms among these
children included vomiting (45%), headache (26.7%), unsteadiness (25%), and paresis
(18.3%). Of the 60 patients, 38 patients had two or more symptoms before diagnosis,
only 16 patients had one symptom before diagnosis, four patients were diagnosed at
23
routine medical check-ups, and two patients were diagnosed during the perinatal period.
A majority of the patients either had headache or vomiting as the first symptom followed
by other common symptoms or more specific neurological symptoms such as cranial
nerve palsies or increased ICP clinical manifestations. Wilne et al. (2012) also found that
between the time of symptom onset and diagnosis, there was an increase in the number of
signs and symptoms (initial onset: ranged from 1-8 symptoms; time at diagnosis: ranged
from 1-17 symptoms). Overall, the initial non-specific signs and symptoms of children
with brain tumors continues to progress to more concerning signs and symptoms such as
paresis, palsies, or increased ICP manifestations.
Klitbo et al. (2011) conducted a retrospective study of 46 children with brain
tumors in Southern Denmark by analyzing medical charts and interviewing parents and
general practitioners. Similar to previous studies, Klitbo et al. (2011) found that the most
common clinical symptom at diagnosis was vomiting (50%), headache (50%), and ataxia
(35%).
Wilne et al. (2012) examined hospital medical records of 139 children diagnosed
with brain tumor in the United Kingdom. The researchers found that the most common
signs and symptoms of children with brain tumors (most common listed first) were
headache, nausea and/or vomiting, motor system abnormalities, cranial nerve palsies,
visual system abnormalities, seizures, endocrine or growth abnormalities, behavior
change, abdominal or back pain, alteration in or loss of consciousness, and spinal
deformity. Over half (56.8%) of these children only had a single symptom at onset,
24
which highlights the non-specificity of these signs and symptoms in relation to a brain
tumor diagnosis.
Wilne et al. (2007) conducted the first known systematic review and meta-
analysis of childhood presentation of CNS tumors to elicit the most common signs and
symptoms of children with brain tumors. The study narrowed the 5,620 potentially
relevant articles down to 74 studies that met the inclusion criteria over a 14-year period
(1991-2005). Sixty-one studies (N = 3,702 children) described signs and symptoms of
intracranial tumors at diagnosis with headaches (33%), nausea and vomiting (32%), and
abnormal gait or coordination (27%) as the most common signs and symptoms. Thirteen
studies (N = 332 children) described signs and symptoms of children with intracranial
tumors under 4-years old with macrocephaly (41%), nausea and vomiting (30%),
irritability (24%), lethargy (21%), abnormal gait and coordination difficulties (19%), and
clinically apparent hydrocephalus with bulging fontanel or splayed sutures (13%) as the
most common signs and symptoms. Eight studies (N = 307 children) described signs and
symptoms of children with neurofibromatosis and intracranial tumors with reduced visual
acuity (41%), exophthalmia (16%), optic atrophy (15%), and squint (13%) as the most
common signs and symptoms. It is important to note that the study had some limitations
including publication bias since unpublished papers were not searched, recall bias due to
data accuracy depended on patient and family history during healthcare visits, and
majority of studies were completed at one center and only four studies included samples
from national trials. Overall, the study provided an estimate of the frequency and types
of the signs and symptoms seen in children with brain tumors.
25
The importance of knowing the most common signs and symptoms that children
may present with to healthcare providers has been recognized by ongoing research.
Without this knowledge of the clinical presentation of childhood brain tumors, healthcare
providers may lack pertinent knowledge and childhood brain tumors will continue to be
misdiagnosed or have prolonged symptom intervals.
Common Misdiagnoses
There was only one study that mentioned common misdiagnoses since
misdiagnoses was not the primary or secondary research question for most studies
included in this review of literature (Dörner et al., 2007). Dörner et al. (2007) noted that
some of the patients were misdiagnosed with common differential diagnoses such as
gastrointestinal infection, appendicitis, psychological behavior problems, cervical spine
strains, or ophthalmologic entities. Though other studies did not mention the common
differential diagnoses that healthcare providers considered, some studies did include the
types of specialists seen before a diagnosis of brain tumor was reached, which will be
mentioned below in this review (Dörner et al., 2007; Edgeworth et al., 1996; Mehta et al.,
2002; Thulesius et al., 2000; Wile et al., 2012).
Tumor Histology and Tumor Location Association with Diagnostic Delay and Signs
and Symptoms
Generally, the type and location of tumor correlates with how early or late a brain
tumor is diagnosed and with the child’s clinical manifestations (Pollack, 1999).
Supratentorial tumors and infratentorial tumors will present differently in children with
brain tumors. Similarly, benign tumors and malignant aggressive tumors will also
26
present differently in children (Pollack, 1999). The diagnosis of a child with a brain
tumor is complicated due to the child’s age and developmental level, especially a younger
child because he/she is unable to express their symptoms verbally; the variety of clinical
presentations depending on histology and location of the tumor, which may potentially
represent other diagnoses; and parental and healthcare provider knowledge of the child’s
medical history, history of present illness, and common signs and symptoms of pediatric
brain tumors (Kukal et al., 2009; Pollack, 1999; Wilne et al., 2010). A specific guideline,
such as the The Brain Pathways Guideline: A Guideline To Assist Healthcare
Professionals In The Assessment Of Children Who May Have A Brain Tumour (The
Brain Pathways Guideline) should provide the healthcare provider with specific steps and
guidance to facilitate the diagnosis of a child with a brain tumor (Walker et al., 2008;
Wilne et al., 2010).
Hyashi et al. (2010) examined medical records of children with brain tumors
retrospectively. The researchers found that vomiting and headache were associated with
every tumor location and that increased ICP clinical manifestations were seen in posterior
fossa tumors (88.2%), supratentorial tumors (69.2%), central tumors (58.8%), uncertain
tumors (50%), and brainstem tumors (18.2%). The aforementioned signs and symptoms
progression are associated with brain tumor location; however, the researchers did not
specify which type of tumor has a more rapid progression of headache and vomiting. The
authors did not examine the tumor histology in correlation to diagnostic delay or signs
and symptoms.
27
Kukal, Dobrovoljac, Boltshauser, Ammann & Grotzer (2009) examined medical
charts retrospectively of pediatric brain tumor patients in Switzerland. Tumor histology
correlated with the PSI meaning that aggressive and fast-growing tumors had shorter PSI
than slow-growing tumors (Kruskal-Wallis statistic=36.09, p
28
and signs and symptoms. High-grade tumors were significantly associated with a shorter
symptom interval. A shorter than median symptom interval was associated with the
initial presentation of nausea and/or vomiting (odds ration [OR]=1.8, p=0.003), abnormal
gait (OR=2.3, p=0.001), coordination difficulties (OR=2.7, p=0.006); focal motor
weakness (OR=2.8, p=0.002), and facial weakness (OR=3.1, p=0.03). A longer than
median symptom interval was associated with initial presentation of head tilt (OR=0.4,
p=0.006) and cranial nerve palsies (OR=0.6, p=0.025).
In conclusion, tumor histology and location plays a role in how early or late
childhood brain tumors are diagnosed and also contributes to the clinical presentation of
the child (Hyashi et al., 2010; Kukal et al., 2009; Reulecke et al., 2008; Wilne et al.,
2012). High-grade tumors are diagnosed sooner than low-grade tumors and aggressive
and malignant tumors are diagnosed earlier than benign tumors due to the tumor
histology and clinical presentation of the child (Kukal et al., 2009; Pollack, 1999; Wilne
et al., 2012). The knowledge and understanding of these important concepts will be of
utmost importance in assisting healthcare providers to recognize the signs and symptoms
of a child suspected of a brain tumor earlier. The use of The Brain Pathways Guideline:
A Guideline To Assist Healthcare Professionals In The Assessment Of Children Who May
Have A Brain Tumour (The Brain Pathways Guideline) can support the healthcare
provider with an easy to access and use tool to assist in the differential diagnosis of
pediatric brain tumors (Walker et al., 2008; Wilne et al., 2010).
29
Diagnostic Delays and Pre-Diagnostic Symptom Interval (PSI)
In an attempt to concentrate efforts in detecting pediatric brain tumors earlier,
researchers have conducted studies to identify where the delays are occurring (i.e. parent
versus doctor), how long the delays are, and compare or correlate the PSI with patient
characteristics (Dobrovoljac et al., 2002; Edgeworth et al., 1996; Flores et al., 1986;
Hyashi et al., 2010; Klitbo et al., 2011; Kukal et al., 2009; Mehta et al., 2002; Pollock et
al., 1991; Thulesius et al., 2000; Wilne et al., 2012; Wilne et al., 2006). Each of these
will be addressed in the following sections.
Pre-Diagnostic Symptom Interval (PSI)
Pre-diagnostic symptom interval refers to the time from initial onset of signs and
symptoms to official diagnosis of a brain tumor by neuroimaging or biopsy (Dobrovoljac
et al., 2002; Edgeworth et al., 1996; Flores et al., 1986; Hyashi et al., 2010; Klitbo et al.,
2011; Kukal et al., 2009; Mehta et al., 2002; Pollock et al., 1991; Thulesius et al., 2000;
Wilne et al., 2012; Wilne et al., 2006). The PSI has been one of the main outcome
measures for majority of studies in this literature review
Dobrovoljac et al. (2002) found that the median PSI of all patients was 60 days
and only 32% of the patients were diagnosed within 30 days after initial presentation of
signs and symptoms. Age was significantly correlated with the PSI (Pearson’s
correlation r=0.32, p
30
technologies. Edgeworth et al. (1996) found a mean PSI of 20 weeks (SD=29.1,
range=
31
(1996) reported mean parental delay of 3 weeks (SD=13.4, range=0-104 weeks). Klitbo
et al. (2011) described a median parental delay of 7 days (range=0-365 days). Kukal et
al. (2009) reported a median parental delay of 14 days (range=0-1,835 days) with
significantly shorter parental delays in younger children (Jonckheere-Terpstra [JT]
statistic=2.23, p=0.025).
Doctor’s Delay
Doctor’s delay refers to the interval from the first consultation to the diagnosis by
neuroimaging technology (CT or MRI). Dobrovoljac et al. (2002) found that the median
doctor’s delay was 30 days and doctor’s delay did not significantly correlate with the
child’s age. Edgeworth et al. (1996) reported mean doctor’s delay of 16 weeks
(SD=24.4, range=0-103 weeks). Klitbo et al. (2011) found a median doctor’s delay of 15
days (range=0-730 days). Kukal et al. (2009) reported a median doctor’s delay of 14
days (range=0-3,480 days) that was not significantly associated with the age of the child
at diagnosis (Jonckheere-Terpstra [JT] statistic=1.5, p=0.13).
Summary of Diagnostic Delays and Pre-Diagnostic Symptom Interval (PSI)
Although availability of neuroimaging technology has increased, the PSI remains
relatively similar to older studies and continues to be unacceptably long. Although the
literature suggests a decreased PSI may lead to detection of smaller brain tumors that are
easier to resect thus resulting in better patient outcome, more research is needed on this
particular matter (Dobrovoljac et al., 2002; Klitbo et al., 2011; Mehta et al., 2002;
Reulecke et al., 2008). From the research studies that did report parental delay and
doctor’s delay, there is an assumption that diagnostic delay can occur at every stage
32
(Dobrovoljac et al., 2002; Edgeworth et al., 1996; Klitbo et al., 2011; Kukal et al., 2009).
Hence, adequate education is not only important for healthcare providers, but also parents
or other family members involved in the child’s care. In other subspecialties, screening
tests are done to allow for earlier detection of illnesses (Mehta et al., 2002), such as
phenylketonuria (PKU), cystic fibrosis (CF), congenital heart defects, and many more
(Centers for Disease Control and Prevention, 2014); the pertinent question for children
suspected of brain tumors is: does this population possibly require screening tests or
tools. From the data presented about the PSI, it appears that screening tests, tools, or
guidelines should be implemented.
33
TABLE 1. Summary of Pre-Diagnostic Symptom Interval (PSI), Parental Delay, and
Doctor’s Delay From Multiple Research Studies
Research Study (Author, Year)
PSI Parental Delay Doctor’s Delay
Dobrovoljac et al. (2002)
Median = 60 days Median = 14 days Median = 30 days
Edgeworth et al. (1996)
Mean = 20 weeks (140 days)
Range =
34
Note: Presence of dashes (–) means that these values were not reported in the study. Text
in parentheses means that the values were converted to days for comparison between all
studies (1 month = 30 days; 1 week = 7 days).
Primary Care Providers Versus Specialist Doctors
Diagnostic delay and lengthy PSI has been established as problematic in pediatric
brain tumor patients from conducting retrospective studies (Dörner et al., 2007;
Edgeworth et al., 1996; Mehta et al., 2002; Wilne et al., 2012). Questions about whether
the diagnostic delay is occurring with primary care providers or specialty doctors and the
frequency of consultation before diagnosis have been mentioned. Some studies have
taken this into consideration when conducting the retrospective studies of children with
brain tumors; however, the studies included in this review of literature do not
differentiate the timing of diagnosis among the different physicians and whether seeing a
primary care provider versus a specialist resulted in earlier or later diagnosis (Dörner et
al., 2007; Edgeworth et al., 1996; Mehta et al., 2002; Wilne et al., 2012). The studies
included address the fact that children with brain tumors see primary care providers and
specialists as well as request consultations more frequently than children without brain
tumors (Dörner et al., 2007; Edgeworth et al., 1996; Mehta et al., 2002; Wilne et al.,
2012). In one study, patients were seen by two other physicians of different specialties
on average before admission into the hospital department (Dörner et al., 2007). These
specialties included pediatrics (hospital), pediatrics (praxis), ears, nose, throat (ENT),
general practitioner, ophthalmologist, neurologist, and orthopedics (Dörner et al., 2007).
In another study, researchers calculated the percentages of different consultations with
35
healthcare providers. Prior to diagnosis, general practitioners were consulted the most
(45.5%), followed by pediatrician (13%), opticians (9%), neurologists (2%), and
psychologists (1%) (Edgeworth et al., 1996). Prior to diagnosis, 63% of all visits were
with primary care providers and 37% were with specialists that included pediatricians,
gastroenterologists, orthopedic surgeons, neurologists, ophthalmologists, and
neurosurgeons (Mehta et al., 2002). Of the 182 children in the study conducted by Wilne
et al. (2012), 81 patients visited the general practitioner, followed by 79 with hospital
pediatrician, 24 with ophthalmologists, 15 with opticians, and 29 with emergency
department doctors. The range of primary care providers and specialists consulted by
pediatric patients with brain tumors highlights the need for all practitioners to have
knowledge of brain tumor presentations because these children did not only present to
primary care providers, but also many different specialist doctors. All of the studies
included in this review of literature did not differentiate between whether the general
practitioners or primary care providers were Doctor of Medicine (MD), Doctor of
Osteopathic Medicine (DO), or Nurse Practitioners (NP).
Number of Consultations or Visits to Healthcare Providers Prior to Diagnosis
Some researchers have noted that children diagnosed with brain tumors have
more visits to doctor’s offices than children without brain tumors (Ansel et al., 2010;
Edgeworth et al., 1996; Mehta et al., 2002; Wilne et al., 2012). Edgeworth et al. (1996)
reported a mean number of 4.6 visits with healthcare providers before diagnosis was
confirmed (45.5% with general practitioner and 9% with an accident and emergency
department). In one study, the number of visits to healthcare providers ranged from 0 to
36
more than 10 where 41% of patients received diagnosis within 3 visits (Mehta et al.,
2002). Wilne et al. (2012) analyzed the number of visits a step further and reported that
patients with longer than average symptom interval were significantly associated with
increased number of healthcare visits. In a retrospective study conducted in the United
Kingdom by Ansel et al. (2010), the number of general practitioner visits was compared
between case and control children. Each case (child diagnosed with brain tumor) was
matched with two controls by gender, month and year of birth, and region of residence.
Case children saw a general practitioner more than controls (31.5 times compared to 25.6
times; 1.22 times more often, 95% CI=1.09-1.36) (Ansel et al., 2010). If a healthcare
provider consistently sees the same child for similar signs and symptoms as previous
visits, then the assumption should be that more examination or diagnostic tests are needed
to evaluate the child appropriately (Ansel et al., 2010; Edgeworth et al., 1996; Mehta et
al., 2002; Wilne et al., 2012).
Diagnostic Delay of Pediatric Brain Tumors in Comparison to Other Childhood
Cancers
Two studies included in this review of literature investigated the PSI of children
with brain tumors compared to other childhood cancers such as leukemia, Hodgkin and
non-Hodgkin lymphoma, and bone tumor groups (Pollock et al., 1991; Thulesius et al.,
2000). According to Pollock et al. (1991), the median PSI for brain tumors was 31 days
compared neuroblastoma of 21 days, non-Hodgkin lymphoma of 26 days, Hodgkin
lymphoma of 49 days, osteosarcoma of 56 days, and Ewing sarcoma of 72 days. Overall,
the longest PSI was seen in Hodgkin lymphoma and bone tumor groups with Ewing
37
sarcoma at the top of the list. In another study conducted by Thulesius et al. (2000), a
comparison of diagnostic delay between children diagnosed with leukemia and brain
tumors was conducted. The median PSI for children diagnosed with leukemia was 3
weeks (range 0-15 weeks) and 9 weeks (range 1-199 weeks) for children with brain
tumors. The parental delay (χ2=9.59, p=0.002) and doctor’s delay (χ2=5.50, p=0.019)
was significantly shorter for leukemia patients compared to brain tumor patients. The
overall PSI was significantly less for leukemia patients compared to brain tumor patients
(χ2=9.52 p=0.002). The two studies present conflicting views about the diagnostic delay
of brain tumors in comparison to other childhood cancers. A conclusion that most
childhood cancers are diagnosed late can be made, but the two studies are not strong
enough to indicate that brain tumors are diagnosed significantly later than other
childhood cancers.
Cost-Effectiveness of Computed Tomography (CT) and Magnetic Resonance
Imaging (MRI) to Diagnose Pediatric Brain Tumors
Children suspected of having a brain tumor are typically referred to places to
obtain imaging of the brain. Because signs and symptoms of pediatric brain tumors are
non-specific, healthcare providers may feel hesitant between ordering neuroimaging
studies with risks of sedation, use of contrast, and false-positive rates and detection of
brain tumor (Medina, Kuntz, & Pomeroy, 2001). Medina et al. (2001) conducted a cost-
effectiveness analysis and quality adjusted life years (QALY) study of three diagnostic
strategies for children with headaches (MRI, CT followed by MRI with positive results
[CT-MRI], and no neuroimaging with close monitoring and follow-up). The sample of
38
children with headaches were divided into three groups: low-risk (non-migraine
headaches for >6 months and normal neurologic exam), intermediate-risk (migraine
headache and normal neurologic exam), and high-risk (headaches for $1 million (US dollars) per QALY
saved compared with no imaging. For the high-risk group with a baseline prevalence of
4%, MRI was the most effective diagnostic strategy because it was less costly than CT-
MRI strategy due to the CT false-positive rate leading to unnecessary workup with MRI.
MRI compared with CT-MRI and no imaging maximized QALY gained meaning that it
was less expensive per QALY when the probability of having a brain tumor increased.
Overall, the MRI strategy in high-risk children with headache maximizes QALY
expectancy at a reasonable cost-effectiveness ratio. MRIs tend to be more expensive than
CTs, but as the cost difference lessened, MRIs became more cost-effective than CTs
because MRIs are more effective at diagnosing brain tumors. The researchers further
suggest a decision tree for use in children with headaches (See Figure 2). This decision
39
tree may be applied to children with other signs and symptoms in addition to headache or
without headache that indicate a space-occupying lesion such as a brain tumor.
FIGURE 2. Decision Tree for Children With Headache (from Medina et al., 2001)
Does Prolonged Pre-Diagnostic Symptom Interval (PSI) Decrease Survival?
Healthcare providers and researchers have the assumption that diagnosing
pediatric brain tumors earlier will increase survival and decrease neurological deficits
(Dobrovoljac et al., 2002; Dörner, et al., 2007; Edgeworth et al., 1996; Flores et al., 1986;
Klitbo et al., 2011; Pollock et al., 1991; Reulecke et al., 2008; Thulesius et al., 2000;
Wilne et al., 2012; Wilne et al., 2007; Wilne et al., 2010). Parents also often question
whether an earlier diagnosis could have been accomplished and whether the diagnostic
delay affects the child’s prognosis adversely (Dixon-Woods et al., 2001; Kukal et al.,
2009). Unfortunately, there is only one current study that has considered whether or not
dictors of a surgical brain tumor such as an abnormalneurologic examination.7,15 Changes in our approachto health care emphasize the need to consider long-term outcomes and costs in evaluating medical diag-nostic strategies. Accordingly, we performed a deci-sion analytic Markov model and CEA of 3 riskgroups incorporating the quality-of-life adjustmentsand costs associated with early versus late braintumor diagnosis and treatment.
The low-risk group with baseline pretest probabil-ity of 0.01% (1/10 000) consisted of patients with thesole finding of chronic headaches of !6 months’duration.25,26 The most effective strategy was no im-aging with close clinical follow-up. Furthermore, thisstrategy was cost-saving when compared with theCT-MRI and MRI strategies because of its greatereffectiveness at a lower cost. Therefore, the risks offalse-positive neuroimaging results, contrast use,and sedation outweigh the benefit of early diagnosisof a rare brain tumor in this risk group (Table 4).
The intermediate-risk group with baseline pretestprobability of 0.4% (4/1000) consisted of childrenwith migraine headaches and a normal neurologicexamination.14,25 CT-MRI was a slightly more effec-tive diagnostic strategy than MRI alone. However,the cost-effectiveness ratio of the CT-MRI strategywas !$1 million per QALY saved.
The high-risk group with baseline prevalence (pri-or probability) of 4% consisted of children with head-ache of "6 months’ duration and at least 1 otherclinical predictor suggestive of a surgical brain tu-mor such as an abnormal neurologic examination.15MRI maximized QALY gained compared with CT-MRI or no imaging in this risk group. An increase inthe number of predictors has been highly correlatedwith an increased risk of brain tumor.15 Assessment
of this variable revealed a decrease in the cost-effec-tiveness ratio from $113 800 to $67 000 per QALYsaved when the prior probability was increased from4% to 8% (Table 4).
The cost-effectiveness ratios allow comparisonswith alternative health care programs and may assistin resource allocation decisions.23 The high-riskgroup cost-effectiveness ratios compared favorablywith other well-accepted diagnostic procedures. Forexample, annual mammography for women 55 to 64years old costs $110 000 per LY saved (updated to1993 US dollars),38 annual cervical cancer screeningfor women beginning age 20 costs $220 000 per LYsaved (updated to 1993 US dollars),38,39 and colonos-copy for colorectal cancer screening for people olderthan 40 years costs $90 000 per LY saved (updated to1993 US dollars).38,40
In the high-risk group we found our analysis to besensitive to the percentage of favorable prognosistumors that progressed to unfavorable prognosis be-cause of delayed diagnosis and treatment, and in-creased risk of death from brain herniation. That is,the incremental cost-effectiveness ratio for MRI waslower for patients who had a greater probability ofprogression because of delayed tumor diagnosis andtreatment, and greater risk of dying from brain her-niation (Table 5).
We also found in our high-risk group that thediagnostic performance of the neuroimaging testsaffected our base case-results. MRI became a domi-nated strategy if its specificity was "97%. CT was nolonger dominated if its specificity was greater than97%. A decrease in the MRI test sensitivity revealedan increase in the cost-effectiveness ratio of the MRIstrategy (Table 5). Therefore, the exact diagnosticperformance (sensitivity and specificity) of MRI and
Fig 2. Suggested decision tree for usein children with headache. For patientsin the high- and intermediate-riskgroup, neuroimaging is suggested. Forpatients in the low-risk group or thosewith negative findings from imagingstudies, clinical follow-up with peri-odic reassessment is recommended.
ARTICLES 261 at Univ Of Arizona on August 20, 2013pediatrics.aappublications.orgDownloaded from
40
diagnostic delay results in decreased survival in this patient population (Kukal et al.,
2009). The researchers calculated 5-year and 10-year overall survival rates (OS) and 5-
year and 10-year progression free survival rates (PFS) for their sample size of 315
patients. The OS was defined as the survival probability with death as the only event and
the PFS was defined as “the probability of being alive and free of progression/relapse;
death and progression/relapse were considered as events” (p. 304). The 5-year OS was
73% (95% confidence interval [CI]=68%-79%) and 10-year OS was 71% (95% CI=66%-
77%). The 5-year PFS was 53% (95% CI=47%-59%) and 10-year PFS was 48% (95%
CI=42%-55%). The patients were categorized into groups according to PSI measured in
days (PSI >180 days group, PSI 60-179 days group, PSI 20-59 days group, and PSI 180 days had a higher survival probability, followed by
PSI
41
TABLE 2. Overall Survival Probabilities (OS) and Progression Free Survival
Probabilities (PFS) in Pediatric Brain Tumor Patients (Kukal et al., 2009)
PSI Groups 10-Year OS 10-year PFS All Tumors (n=315) χ2=16.4, p
42
care. Some parents experienced lengthy disputes with their primary healthcare providers
and secondary healthcare providers such as specialty doctors. Parents in the lengthy
dispute category described healthcare providers as incompetent due to their failure to
recognize the seriousness of the signs and symptoms. They also perceived the healthcare
system or medical response as inadequate because of the delay in appropriate
investigation of their child’s signs and symptoms. After finally obtaining a diagnosis, the
parents felt relieved because at least there was an identifiable cause for the unusual signs
and symptoms and it validated their initial concerns. Trustworthiness of this study may
be questioned due to a small sample size (20 families), inclusion of only one pediatric
oncology unit, and inadequate description of analysis procedures because it is unknown
whether researchers performed investigator triangulation or member checking.
The study conducted by Dixon-Woods et al. (2001) offers valuable insights into
where diagnostic delays may occur in the healthcare system. Parental concerns of their
child are valuable information for healthcare providers and shows that parents play an
important role in improving the PSI. Attention has been given to the roles of healthcare
providers in attempt to improve the PSI, but consideration should also be given to the
roles of parents, children, teachers, school nurses, and other people involved in the
child’s care. Attentiveness to the different roles each group plays may help decrease the
PSI, however, future research studies will need to examine these aspects.
Healthcare Provider Guideline to Assist in Diagnosis of Pediatric Brain Tumors
From the aforementioned research studies and issues addressed, there is an
identified need for a guideline to help healthcare providers assess and diagnose children
43
who may have a brain tumor. Appendix A Figure 1 illustrates The Brain Pathways
Guideline, an evidence-based guideline developed by The Children’s Brain Tumour
Research Centre at Nottingham University to assist practitioners in identifying, assessing,
and investigating children suspected of having a brain tumor (Walker et al., 2008; Wilne
et al., 2010). The guideline is developed from evidence in systematic reviews, meta-
analysis, and cohort studies. In order to translate the research evidence into a clinical
guideline, a multidisciplinary workshop and Delphi process that included healthcare
providers and parents with children diagnosed with brain tumors was assembled to
address key issues of signs and symptoms, specificity, assessment, referrals, and imaging.
The guideline addresses when healthcare providers should consider a brain tumor in a
child (i.e. when the child presents with headache, nausea and/or vomiting). It also
addresses each common sign and symptom separately that are grouped into categories
(headaches, nausea and vomiting, visual symptoms and signs, motor symptoms and signs,
growth and development, and behavior) and recommends referral and imaging pathways
depending on the healthcare provider’s assessment (i.e. imaging required if patient has
persistent vomiting upon awakening) (Walker et al., 2008; Wilne et al., 2010).
The aims of this guideline are for healthcare providers to be aware of pediatric
brain tumor signs and symptoms and offer an understandable guideline to assess and
investigate children suspected of having a brain tumor. Currently, there is no evidence on
whether or not the guideline is effective in decreasing the pre-diagnostic symptom
interval, but the researchers are planning and developing a dissemination program to
implement the guideline. The researchers plan to review the evidence, literature, audit,
44
and feedback from users 5 years after publications to evaluate and update the guideline
(Walker et al., 2008; Wilne et al., 2010). The aims of this DNP Project to assess
healthcare provider knowledge of pediatric brain tumors will help add evidence to the
researchers evaluation of the guideline.
The aim of any guideline is to assist healthcare providers and patients to make
appropriate health care decisions for a specific clinical situation. Guidelines may be
beneficial in the healthcare setting, but are only as good as the evidence that supports the
development of the guideline. Rigorous methodologies and strategies are needed when
developing a guideline that can be implemented successfully in the practice setting.
Guidelines may vary in its quality hence the requirement for an instrument to assess the
quality. The Appraisal of Guidelines for Research and Evaluation (AGREE) Instrument
has been developed to “assess the quality of guidelines; provide a methodological
strategy for the development of guidelines; and inform what information and how
information ought to be reported in guidelines” (Brouwers et al., 2010, p. 1). The
AGREE II Instrument is composed of six domains with 23 items and each item receives a
score from 1 (strongly disagree) to 7 (strongly agree). The six domains include: scope
and purpose (three items), stakeholder involvement (three items), rigor and development
(eight items), clarity of presentation (three items), applicability (four items), and editorial
independence (two items) (Brouwers et al., 2010). The AGREE II Instrument is used by
the primary investigator (AT) to evaluate the quality of the guideline developed by The
Children’s Brain Tumour Research Centre at Nottingham University. The scoring of the
guideline based on the AGREE II Instrument domains are as follows: domain 1 = 21/21
45
(domain score 100%); domain 2 = 17/21 (domain score 82.35%); domain 3 = 43/56
(domain score 72.92%); domain 4 = 19/21 (domain score 88.89%); domain 5 = 25/28
(domain score 87.5%); and domain 6 = 14/14 (domain score 100%). Overall, the quality
of the guideline is rated at a 6 out of 7 though it has a few weaknesses such as the lack of
expert review prior to its publication and minimal evidence from systematic reviews and
meta-analyses.
Synthesis of Literature
Literature shows that children are diagnosed late because healthcare providers
consistently do not recognize the clinical presentation of pediatric brain tumors (Dörner
et al., 2007; Hyashi et al., 2010; Klitbo et al., 2011; Mehta et al., 2002; Reulecke et al.,
2008; Wilne et al., 2012; Wilne et al., 2007; Wilne et al., 2006). There is agreement that
the initial signs and symptoms of children with brain tumors are non-specific and may
resemble other less serious illnesses leading to an exclusion of a brain tumor diagnosis on
the differential diagnosis list (Dörner et al., 2007; Hyashi et al., 2010; Klitbo et al., 2011;
Mehta et al., 2002; Reulecke et al., 2008; Wilne et al., 2012; Wilne et al., 2007; Wilne et
al., 2006). Numerous studies have sought to identify the common signs and symptoms
related to pediatric brain tumors in hopes of increasing awareness among healthcare
providers (Dörner et al., 2007; Hyashi et al., 2010; Klitbo et al., 2011; Mehta et al., 2002;
Reulecke et al., 2008; Wilne et al., 2012; Wilne et al., 2007; Wilne et al., 2006).
Consensus is that the signs and symptoms of brain tumors in children are related to tumor
location, tumor histology, and age of child, but the initial sign and symptom is not
46
specific to a brain tumor diagnosis and fails to alert the healthcare provider (Hyashi et al.,
2010; Kukal et al., 2009; Pollack, 1999; Reulecke et al., 2008; Wilne et al., 2012).
The pre-diagnostic symptom interval (PSI) has not improved dramatically with
the increase availability of neuroimaging technologies (Dobrovoljac et al., 2002;
Edgeworth et al., 1996; Medina et al., 2001). There continues to be concerning
diagnostic delay of childhood brain tumors leading to life-threatening emergencies and
devastating neurological deficits among survivors (Dobrovoljac et al., 2002; Edgeworth
et al., 1996; Flores et al., 1986; Hyashi et al., 2010; Klitbo et al., 2011; Kukal et al., 2009;
Mehta et al., 2002; Pollock et al., 1991; Thulseius et al., 2000; Wilne et al., 2012; Wilne
et al., 2006). Literature demonstrates that diagnostic delay is occurring at multiple levels
of care involving parents, primary care providers, and specialists (Dörner et al., 2007;
Mehta et al., 2002; Wilne et al., 2012). Children with brain tumors often see a doctor
more than children without brain tumors and still healthcare providers are missing these
subtle signs (Ansel et al., 2010; Edgeworth et al., 1996; Mehta et al., 2002; Wilne et al.,
2012). These factors impact the relationship between healthcare providers dramatically
because parents of children with brain tumors question whether or not an earlier
diagnosis was possible and if an earlier diagnosis would lead to better prognosis and less
complications or deficits (Dixon-Woods et al., 2001).
Important considerations of the review of relevant literature are that most studies
were conducted outside of the United States, mainly the United Kingdom. The majority
of studies used a retrospective design which itself has limitations due to recall bias when
parents are giving the child’s medical history or inaccurate documentation by healthcare
47
providers regarding dates of signs and symptoms and official diagnosis. Also,
generalizability of the studies included in this literature review is difficult due to sample
characteristics (children) and setting characteristics (mainly United Kingdom). Even
with these internal and external validity concerns, the literature review is representative
of the diagnostic delay concern in the population of children with brain tumors.
In conclusion, the literature review conducted identifies the need for a guideline
to support healthcare providers in recognizing the signs and symptoms of childhood brain
tumors and investigating further with appropriate referrals and neuroimaging in a timely
manner. A guideline has been developed by The Children’s Brain Tumour Research
Centre at Nottingham University (Walker et al., 2008; Wilne et al., 2010) called The
Brain Pathways Guideline and is identified as appropriate for this DNP Project’s aims to
assess healthcare provider knowledge of pediatric brain tumors with a long-term goal of
evaluating whether a guideline assisting healthcare providers will decrease the pediatric
brain tumor pre-diagnostic symptom interval.
48
CHAPTER III: METHODOLOGY
Study Aims
This DNP Project aimed to assess healthcare provider knowledge of pediatric
brain tumor patients using The Brain Pathways Guideline: A Guideline To Assist
Healthcare Professionals In The Assessment Of Children Who May Have A Brain
Tumour (The Brain Pathways Guideline). While the long-term goal is to evaluate the
effectiveness of the guideline in decreasing the pre-diagnostic symptom interval among
children who may have a brain tumor, this DNP Project was the first step to achieving the
long-term goal by assessing healthcare provider knowledge of pediatric brain tumors.
Model to Implement Guideline: ACE Star Model of Knowledge Transformation
Healthcare continues to demand quality improvements that include evidence-
based practices to achieve effective and safe patient outcomes (Stevens, 2012b).
According to the Institute of Medicine (2001), there continues to be a gap in translational
research where the results of studies are not being clinically practiced in a timely manner.
Many studies present an estimate of 17 years for results from research studies to be
transitioned into clinical practice (Morris, Wooding, & Grant, 2011). In addition,
researchers have noticed that 30-40% of patients receive care that is not consistent with
the most recent scientific evidence (Eccles, Foy, Sales, Wensing, & Mittman 2012). The
need to implement evidence-based results into practice continues to be a general
consensus among healthcare providers and researchers (Stevens, 2012b). Evidence has
shown that using a conceptual model is one of the most effective ways in achieving
implementation of evidence-based practice (EBP) into clinical settings. However, there
49
is uncertainty as to which model is the most effective (Stevens, 2012b). Each model has
its own strengths and weaknesses and the appropriateness of the model depends on the
innovation being considered (Bucknall & Rycroft-Malone, 2010).
This DNP Project used the Academic Center for Evidence-based Practice (ACE)
Star Model of Knowledge Transformation to guide the implementation of The Brain
Pathways Guideline by first assessing healthcare provider knowledge of pediatric brain
tumor patients. The ACE Star Model of Knowledge Transformation “emphasizes crucial
steps to convert one form of knowledge to the next and incorporates best research
evidence with clinical expertise and patient preferences” (Stevens, 2012b, p. 19). The
model uses a five-star point figure to illustrate the steps of knowledge transformation:
Point 1: Discovery Research; Point 2: Evidence Summary; Point 3: Translation to
Guidelines; Point 4: Practice Integration; Point 5: Process, Outcome Evaluation (See
Figure 3) (Stevens, 2012a; Stevens, 2012b).
FIGURE 3. ACE Star Model of Knowledge Transformation (from Stevens, 2012a)
50
Star Point 1: Discovery Research
Discovery research involves generating new knowledge from traditional scientific
methodologies and inquiry. In this stage, knowledge is found by using quantitative and
qualitative methods. This point sets the stage for research studies that answers a variety
of research questions (Stevens, 2012a; Stevens, 2012b).
Star Point 2: Evidence Summary
Evidence summary involves synthesizing primary research studies from Star
Point 1 to meaningful statements or summaries such as systematic reviews. Summaries
may also generate new knowledge by combining results from multiple studies to identify
any gaps in knowledge, detect bias results, and recognize chance effects (Stevens,
2012a).
Star Point 3: Translation to Guidelines
Translation to guidelines aims at forming useful summaries, Clinical Practice
Guidelines (CPGs), for healthcare providers and patients to use in the appropriate
situations. CPGs are considered tools that guide organizations, healthcare providers, and
patients to make informed decisions that are contextualized to a specific patient
population and clinical setting (Stevens, 2012a).
Star Point 4: Practice Integration
Practice integration implicates that individual and organizational practices will be
changed through formal and informal techniques. In order for CPGs to accomplish its
purpose, CPGs must be implemented into practice to benefit patients. This stage requires
implementation of innovations that are discovered and developed from Star Points 1 to 3
51
and may perhaps be viewed as the most familiar and important stage (Stevens, 2012a).
Within this stage, the complexity of behavior change is also considered because many
factors can influence the uptake of evidence-based practices and clinical practice
guidelines. Hence, prior to implementation, assessment of knowledge, barriers, supports,
and stakeholders are necessary to facilitate successful implementation of any CPG or
EBP (Graham & Tetroe, 2007).
Star Point 5: Process, Outcome Evaluation
Evaluation is the final stage in the model and may contribute to the cultivating
knowledge base. Evaluation occurs at multiple endpoints that may focus on
organizations, healthcare providers, and/or patients. The purpose of evaluation is to ask,
“what works and what does not work?” so that adjustments can be made to facilitate
successful implementation and sustain the innovation to benefit patients (Stevens, 2012a).
Applicability of ACE Star Model of Knowledge Transformation to The Brain
Pathways Guideline
Star Points 1-3: The Brain Pathways Guideline
The Children’s Brain Tumour Research Centre at Nottingham University
conducted primary studies, systematic reviews, and multidisciplinary workshops and
Delphi process that accomplished the first three points of the ACE model with the end
product called The Brain Pathways Guideline (Walker et al., 2008; Wilne et al., 2010).
Therefore, this DNP Project did not address the first three star points of the ACE model.
52
Star Point 4 and 5
This DNP Project’s long-term goal is to implement The Brain Pathways
Guideline to evaluate its effectiveness in decreasing the pre-diagnostic symptom interval
among the pediatric brain tumor population. Prior to the achievement of this goal was to
assess healthcare provider knowledge of pediatric brain tumor patients, which was of
utmost importance as this may help guide future research studies pertaining to this topic
so that successful implementation of the guideline can be achieved to benefit patients. In
addition, the process of dissemination of The Brain Pathways Guideline was evaluated.
Methods
A one group pre-test and post-test quantitative approach was used to assess
healthcare provider knowledge of pediatric brain tumor patients. Pre-tests and post-tests
are commonly chosen for behavioral research because it allows researchers to measure
change that resulted from the intervention (Dimitrov & Rumrill, Jr., 2003; Polit & Beck,
2012). For this DNP Project, the intervention was The Brain Pathways Guideline
portrayed in a figure/chart (See Appendix A), a symptom card (See Appendix B), and a
quick reference guide that is a shortened version of the original guideline.
The pre-test and post-test (See Appendix C and D) was developed by the primary
investigator and reviewed by three experts in the area of pediatric brain tumor diagnosis
and pediatric neurology/neurosurgery (Table 3, 4, and 5). The pre-test and post-test
questions were identical; however, the post-test had additional questions to evaluate the
process. The pre-test and post-test had five questions that specifically pertained to The
Brain Pathways Guideline with three distractor questions. Types of questions consisted
53
of demographics; the healthcare provider’s identification of correct signs, symptoms, and
diagnosis; knowledge and skills for additional focused assessment of child; and
healthcare provider’s plan for either follow-up or referral. The pre-test, post-test, and The
Brain Pathways Guideline were conducted via electronic mail (e-mail). Part one of the
implementation included an e-mail that was sent to the participants with only the pre-test.
Part two of the imple