COLORECTAL CANCER SCREENING

DNP Project Final Report
Increasing Colorectal Cancer Screening in the Primary Care Setting
Ashley McKillop, BSN, RN
Arizona State University College of Nursing & Health Innovation
Email: Afmckill@asu.edu

1

COLORECTAL CANCER SCREENING

2

Abstract
Purpose: The purpose of this project was to implement a change in workflow to increase
colorectal cancer (CRC) screening rates and improve Meaningful Use scores in a primary care
setting.
Background and Significance: CRC is the second leading cause of cancer-related deaths in the
United States among men and women. Current CRC screening rates remain low, even with
advanced screening options available. Meaningful Use sets specific objectives for health care
providers to achieve. Documenting CRC screening status and recommending CRC screenings to
patients is one of the objectives of Meaningful Use and is considered a Clinical Quality Measure
(HealthIT.gov). Factors that lead to CRC screening include primary care providers (PCPs)
raising the topic, involving support staff, involving patients in the decision-making process, and
setting alerts in electronic health records (EHRs).
Methods: The Health Belief Model and Ottawa Model of Research Use helped guide this
project. The project took place at a private primary care practice. The focus was on patients
between the ages of 50 and 75 years old meeting criteria for CRC. Five PCPs and five medical
assistants (MAs) chose to participate in the study. Participants were given pre and post Practice
Culture Assessment (PCA) surveys to measure perceptions of the practice culture. The project
included a three-part practice change: PCP and MA education about CRC screening guidelines,
EHR documentation and reminders, and a change of patient visit workflow which included
having MAs review patient's CRC screening status before they were seen by the PCP, and
handing out CRC screening brochures when appropriate. PCPs then ordered the appropriate CRC
screening, and the MA documented the screening in the EHR under a designated location. CRC
Screening Project Evaluation Forms were completed by MAs after each patient visit.

COLORECTAL CANCER SCREENING

3

Outcomes: No significant difference from pre to post survey satisfaction scores were found (t
(8) = -1.542, p = .162). Means of quantitative data were reported from the CRC screening
evaluation forms; N=91. The most common method of screening chosen was colonoscopy, 87%.
A strong correlation was found (r (-.293) = .01, p <.05) between receiving a CRC brochure and
choosing a form of screening. Meaningful Use scores post intervention were inconclusive as data
was not made available by the site.
Conclusions: Patients are more likely to choose a screening method when the topic is raised in a
primary care setting. Continued staff education on workflow is important to sustain this change.
Further research is needed to evaluate cost effectiveness and sustainability of this practice
change.
Keywords: colorectal cancer, primary care, screening, Meaningful Use

COLORECTAL CANCER SCREENING

4

Increasing Colorectal Cancer Screening in the Primary Care Setting
This paper introduces the topic of colorectal cancer, its prevalence in the Unites States,
the importance of cancer screenings, and the ways in which advanced practice nurses can have
an impact on its outcomes.
Colorectal cancer (CRC) is the second leading cause of cancer-related deaths in the
United States among men and women, even with the current screening options available.
Worldwide, CRC is the third most common cancer, affecting approximately 1.4 million people in
2012 (World Cancer Research Fund International, 2017). In 2016, approximately 134,000 people
were diagnosed with CRC, accounting for approximately 49,000 deaths (US Preventive Services
Task Force [USPSTF], 2016). The USPSTF finds that "screening for colorectal cancer is a
substantially underused preventive health strategy in the United States" (US Preventive Services
Task Force, 2016). In addition, one of the goals of Healthy People 2020 is to reduce the number
of cancer-related deaths through early detection and treatment (Healthypeople.gov, 2017). The
USPSTF currently recommends screening for colorectal cancer for both men and women
between the ages of 50 to 75 years. Screenings are recommended for patients that are at average
risk and who are asymptomatic, meaning no family or personal history of CRC, blood in their
stool, iron deficiency anemia, or personal history of irritable bowel disease (IBD) or Crohn's
disease (US Preventive Services Task Force, 2016). According to the Colon Cancer Alliance
(2017), 50% of new CRC cases occur in people age 50 and older. People with first degree
relatives who have had CRC have increased risk, and should be screened earlier. It is
recommended that African Americans have their first screening at the age of 45 years.
Current CRC screening rates remain low, even with advanced screening options
available. It is estimated that only one in three adults receive some form of CRC screening

COLORECTAL CANCER SCREENING

5

between the ages of 50 to 75 years (American Gastroenterological Association, 2017). Colorectal
cancer can be caught early with regular screenings. Early detection and treatment leads to better
patient outcomes and improved mortality. The USPSTF recently changed its statement on the
best way to screen for CRC, stating that any form of screening is better than no screening at all.
They do not list screening tests in any order preference, though colonoscopy remains the gold
standard (Hite and Margolin, 2016). Available screenings methods include colonoscopy, flexible
sigmoidoscopy, computed tomography colonoscopy, the guaiac fecal occult blood test (FOBT),
the fecal immunochemical test (FIT), stool DNA test, and the SEPT9 DNA test (US Preventive
Services Task Force, 2016). These screening options all have their own unique risks and
benefits to their testing. Some can be completed by the patients themselves in the privacy of their
home, while endoscopies must be completed in a surgical setting, often with sedation. Overall,
CRC is a common cancer affecting millions of people in the United States each year. Multiple
guidelines recommend screening for CRC between the ages of 50 to 75 years, and yet screening
rates remain low, leading to increased mortality.
Background and Significance
There are multiple guidelines recommending the routine screening for colorectal cancer (CRC)
in people between the ages of 50 to 75 years old. The conventional screening methods include
invasive tests, such as endoscopy/colonoscopy, as well as stool tests that are noninvasive.
However, there are newer screening options available that have the potential to improve CRC
screening rates by offering noninvasive testing in the home setting (Imperiale et al., 2014; Hite &
Margolin, 2016; Patel & Kilgore,2015). The U.S. Preventative Services Task Force (USPSTF)
(2016) has updated its position statement on routine CRC screenings, stating that any form of
CRC is acceptable and better than no CRC screening being done at all. This gives more options

COLORECTAL CANCER SCREENING

6

to providers and allows the opportunity to offer noninvasive screening tests in addition to the
standard invasive colonoscopy procedure.
Hagoel, Neter, Stein, and Rennert (2016) found that in order to increase CRC screenings,
the topic of CRC must be posed by the provider. This sounds simple and even obvious, but it is
not always being carried out in practice. If screening options and prompts are not provided by
primary care providers routine CRC screening rates will not improve. Primary care providers
face many barriers to recommending CRC screenings, including lack of time, logistics,
inconvenience, and provider forgetfulness (Spruce and Sanford, 2012). System barriers include
lack of health insurance, no reminder systems, lack of time, and no access to colonoscopy
(Spruce & Sanford, 2012). Dolan, Boohaker, Allison, and Imperiale (2013) found that patient
preference of screening tests is an important factor in deciding which screening test, if any, to
complete. Each testing option comes with limitations, risks, and benefits. With colonoscopy,
there is risk of post procedure bleeding, rupture, and infection. Colonoscopy also requires a
bowel preparation, transportation to and from the testing facility, time away from work, and
frequently sedation. These risks should be considered and discussed with patients thoroughly
when recommending them for screening (Bibbins-Domingo et al., 2016). Stool based testing,
pose less risks, but a risk includes false positive results that can lead to unnecessary follow-up
testing (US Preventive Services Task Force, 2016). This method also requires patients to handle
their own stool, which could be a deterrent for some. In addition, there is the opportunity for
patients to turn in samples from other family members, while claiming it as their own to avoid
further testing.
There are multiple studies showing multitarget stool DNA (mt-sDNA) tests having
increased sensitivity in detecting CRC compared to fecal immunochemical test (FIT) testing

COLORECTAL CANCER SCREENING

7

(Fitzpatrick-Lewis et al., 2016; Imperiale et al., 2014; Malik, 2016). In 2014 the US Food and
Drug Administration approved the first mt-sDNA test under the brand name Cologuard. Its use is
gaining wide popularity among primary care physicians and is now covered through Medicare.
With Medicare offering coverage for this test, it is believed that private insurance companies will
soon be covering it as well. Patel & Kilgore (2015) found any form of CRC screening to be more
cost effective than having no screening at all, by using incremental cost-effectiveness ratios
(ICERs). The highest ICER was for stool DNA testing every 5 years. However, it is still unclear
as to which screening method is most cost effective. Cologuard's sensitivity and specificity in
detecting CRC is 92% and 84%. High-sensitivity fecal occult blood test (FOBT) has a sensitivity
ranging from 62-79% and a specificity ranging from 87% to 96% (US Preventive Services Task
Force, 2016). Overall, the USPSTF acknowledges that there are multiple screening tools
available, each with their own advantages. Their overall position statement is "the best screening
tests is the one that gets done" (US Preventive Services Task Force, p. 2573, 2016).
There are many factors and barriers that impact people receiving CRC screening during
the recommended age range. Many studies have found that a team-based approach involving
clinic staff can help increase CRC screening rates (Basch, 2015; Baxter, 2017; Potter et al.,
2009). Healthcare professionals are aware of the importance of CRC screenings and the impact
it has on people's health. However, CRC screening rates remain low and thousands of people die
each year from CRC. Previous successful approaches to this problem include using multiple
strategies to increase screenings, involving nonclinical staff, and incorporating a shared decision
model with patients so they can make the best-informed decision (Schroy et al., 2016; Potter et
al., 2009).

COLORECTAL CANCER SCREENING

8

Colorectal cancer is the second leading cause of cancer-related deaths in the United
States among men and women, even with the current screening options available. Worldwide,
CRC is the third most common cancer, affecting approximately 1.4 million people in 2012
(World Cancer Research Fund International, 2017). In 2016, approximately 134,000 people were
diagnosed with CRC, accounting for approximately 49,000 deaths (USPSTF, 2016). The
USPSTF finds that "screening for colorectal cancer is a substantially underused preventive health
strategy in the United States" (USPSTF, 2016). In addition, one of the goals of Healthy People
2020 is to reduce the number of cancer-related deaths through early detection and treatment
(Healthypeople.gov, 2017). The USPSTF currently recommends screening for colorectal cancer
for both men and women between the ages of 50 to 75 years. Preventative screenings are
recommended for patients that are at average risk and who are asymptomatic, meaning no family
or personal history of CRC, blood in their stool, iron deficiency anemia, or personal history of
irritable bowel disease (IBD) or Crohn's disease (USPSTF, 2016). According to the Colon
Cancer Alliance (2017), 50% of new CRC cases occur in people age 50 and older. People with
first degree relatives who have had CRC have increased risk and should be screened earlier. It is
recommended that African Americans have their first screening at the age of 45 years.
Current CRC screening rates remain low, even with advanced screening options
available. It is estimated that only one in three adults receive some form of CRC screening
between the ages of 50 to 75 years (American Gastroenterological Association, 2017). Colorectal
cancer can be caught early with regular screenings. Early detection and treatment leads to better
patient outcomes and decreased mortality. Multiple guidelines recommend the first screening for
CRC between the ages of 50 to 75 years. Follow up screenings after initial screenings vary based
off findings from the first colonoscopy and family history.

COLORECTAL CANCER SCREENING

9

Problem Statement & Picot
In a primary care setting in the Southwest United States, CRC screening
recommendations are being offered at annual visits by primary care providers, if a patient present
with symptoms, or if the patient brings up the topic. There is no unified or team approach to
CRC screenings. There is also a lack of follow up by primary care providers after the
recommendations are given. Often the provider's only way of knowing if the CRC screening was
completed is when a report is sent by the gastroenterologist after the visit. This places the
responsibility of completing the screening on the patient alone. This assessment has led to the
clinically relevant PICOT question, in patients between the age of 50 and 75 years old requiring
screening for colorectal cancer (P), how does offering stool DNA testing and FIT testing in
addition to a screening colonoscopy (I), compared to offering a screening colonoscopy alone (C),
affect the number of patients completing their colorectal cancer screening (O), over a two-month
period (T)?
Search Strategy
To answer this clinically relevant question, an exhaustive search of the literature was
performed. Databases searched included Cumulative Index of Nursing and Allied Health
Literature (CINAHL), PubMed, and The Cochrane Library.
CINAHL
The key term colorectal cancer screening for a yield of 2,775 citations. Then methods of
colorectal cancer screening was searched for a yield of 178 papers. Combining these terms with
the Boolean connector AND resulted in 169 citations. Then the key search terms fecal
immunochemical, FIT, colonoscopy, flexible sigmoidoscopy, guaiac fecal occult blood test, and
stool DNA were searched independently and then combined with the OR connector for a total

COLORECTAL CANCER SCREENING

10

yield of 8276 papers. The terms primary care were then searched independently and then
combined for a total of 85,436 citations. When all major concepts were then combined, there was
a total yield of 427 citations. Further limitations included studies completed in the United States
or Canada only, adult populations age 50-75 years old and studies completed between 2009-2017
for a yield of 66 studies. Rapid critical appraisals were performed using a standardized checklist
form on these remaining studies with 7 being retained for further evaluation.
PubMed
Initial search phrase using PubMed included colorectal cancer screening, stool DNA, and
FIT for a yield of 32 results. Additional criteria of studies between the year 2009-2017 were
added to yield 30 results. Limiting studies to clinical trials only yielded 2 results, therefore the
limitation was removed for a total of 30 studies. The terms primary care and practitioner were
searched independently within these studies. After a rapid critical appraisal two were retained for
further evaluation.
The Cochrane Library
Lastly, a search using The Cochrane Library was performed with key terms colorectal
cancer screening which had an initial yield of 1,962 results. Additional key terms were added to
include stool DNA, FIT, and primary care. The term primary care yielded too many results at 55,
690 results. Ultimately this term was removed to yield a final of 7 studies.
After rapid critical appraisals, ten studies were retained for further evaluation (Appendix
H). These studies were chosen based on their relevance to the PICOT question, their level of
evidence, and quality of their study. Those that were excluded were of low quality, poor study
design, or were not relevant.

COLORECTAL CANCER SCREENING

11

Evidence Synthesis
The ten chosen studies were further critically evaluated and were entered into an
evaluation table (Appendix A) to better allow for synthesis of evidence. The studies include six
randomized control trials (RCT), one survey, two systematic review and meta analyses, and one
cross sectional study. Varying levels of evidence were included among the studies, six of which
were high level I, one level IV, two level V, and one level VI. Large samples sizes were included
in all the studies. One study included a large degree of bias due to the funding of the study being
the maker of the only stool DNA test available on the market currently. This study was still
included due to the high sensitivity and specificity of the test in detecting CRC and its innovative
testing capability. All studies included CRC as its main area of focus, with all but three having a
positive impact on CRC screening status. All but two studies approached screening for CRC in
different ways. There were a wide variety of measurement tools used, likely due to various
healthcare settings, resources, and different healthcare systems. Measurements consisted of CRC
screening status completion rates, intent to screen, or provider satisfaction with available
screening tools. Validity and rate of detection for polyps were included in one study due to the
nature of the study in evaluating FIT and stool DNA tests. Two studies evaluated physician's
roles and involvement in CRC screening and took this into consideration with their results. Three
studies included a multidisciplinary team approach and its effect on CRC screening status. Only
one study cited a specific theoretical framework to guide its design.
A high degree of homogeneity was identified with regards to the study population and
demographics. This is due to the population most commonly affected by CRC and recommended
screening guidelines by the USPSTF. There was moderate heterogeneity regarding populations
having health insurance, being an underserved population, and ethnicity. There was a moderate

COLORECTAL CANCER SCREENING

12

degree of heterogeneity in regard to interventions implemented, due to the wide variety of CRC
screening tools available. While there was high homogeneity regarding outcomes measured,
being CRC screening status, some studies also measured physician involvement, support staff
involvement, patient reminders, and the use of decision aid tools. The most common CRC
screening tools studied were colonoscopy, FIT, and gFOBT. This synthesis of evidence and
current CRC screening guidelines supported the PICOT question a provided reasoning for
implementing a project change that would increase CRC screening rates (Appendix B).
Purpose Statement
The purpose of this project was to implement a change in workflow to increase colorectal
cancer screening rates and improve Meaningful Use scores in a primary care setting.
Meaningful Use was introduced by the United States government in 2009 as part of the
Health Information Technology for Economic and Clinical Health Act (HITECH). The goal of
Meaningful Use was to encourage health care providers to adopt electronic health records system
to improve the quality, safety, and efficiency of patient care. Meaningful Use sets specific
objectives for health care providers and hospitals to achieve to qualify for Centers for Medicare
and Medicaid Services Incentive Programs. Documenting colorectal cancer screening status and
recommending CRC screenings to patients is one of the objectives of Meaningful Use and is
considered a Clinical Quality Measure (HealthIT.gov).
Evidenced Based Practice Model And Conceptual Framework
The Health Belief Model (HBM) (Appendix C) guided the proposed project design. The
HBM was developed in the 1950's and explores why people do not partake in health-screening
programs (Ueland, Hornung, and Greenwald, 2006). This model has been used in multiple
studies to explore CRC screening uptake. The HBM has six key concepts; perceived

COLORECTAL CANCER SCREENING

13

susceptibility, perceived severity, perceived benefits, perceived barriers, cues to action, and selfefficacy (Ueland, Hornung, & Greenwald, 2006). The HBM was chosen for this practice change
because CRC is considered a preventable cancer if caught early on with CRC screenings.
Patient's belief of a perceived threat, CRC, and perceived benefits of screening, can improve
outcomes and promote self-efficacy when they decide to have CRC and within the recommended
timeframe.
The Ottawa Model of Research Use (Appendix D) was used as the Evidence Based
Practice (EBP) to facilitate practice change. The Ottawa Model consists of six steps that are used
to guide research into practice (Logan, Harrison, Graham, Dunn, and Bissonnette, 1999). The
Ottawa Model is widely used in research and is recognized for its ability to implement EBP. The
structured approach of assessing barriers and facilitators, monitoring interventions, and
evaluating outcomes was the reason this model was chosen. The Ottawa Model takes the patient,
practitioner, and environment into consideration to implement a change in practice, all are
important factors when it comes to CRC screening. This model provided a systematic and
collaborative approach to adopting changes in the way the practice approaches CRC screening.
Project Methods
Ten primary care physicians (PCPs) and their medical assistants (MAs) at the designated
clinic were offered to participate in this evidence-based practice change project. The participants,
primary care providers, DNP student, and medical assistants, were expected to spend
approximately 8-10 weeks participating in the practice change, or until up to 300 patients had
been offered screening. Ultimately, five PCPs and five MAs chose to participate in this practice
change. Project data collection took place between October 2017 and February 2018.

COLORECTAL CANCER SCREENING

14

Institutional review board (IRB) approval was received on September 7th, 2017 and
granted by Arizona State University. The DNP student then met with PCPs, MAs, and the office
manager to begin the project. Information on the project design, goals and purpose of the
practice change were discussed with individuals during one-on-one meetings at the clinic site.
An information sheet about the project was provided at these one-on-one meetings. The Practice
Culture Assessment (PCA) survey (Appendix E) was used to measure perceptions of the practice
culture. The PCA is a 22-question tool designed to measures perceptions related to practice
function and successful implementation of practice quality improvement; permission to use this
survey was granted from one of its authors Dr. Perry Dickinson. The PCA has been used in
multiple studies to evaluate workplace change (Dickinson et al., 2015; Jaén et al., 2010). This
survey acted as a pre and post survey that was given at the beginning of the project and when the
project was finished. The attendance at the educational meeting and completion of the PCA
survey was used as consent to participate in the practice change. Participants were given a unique
subject ID code to match pre and post surveys; all participants responses remained anonymous.
Once participants were recruited to participate, the practice change began. Medical assistants
identified patients that meet criteria for CRC screening based on the education they received
during the education meeting and they were instructed to follow a CRC Screening Flowsheet
(Appendix H) provided to them. If eligible patients did not have any form of CRC screening
documented in their electronic health record (EHR), stated they had not had a screening, or were
due for repeat screening, the medical assistant handed patients the CDC Designed Colon Cancer
Educational Brochure (Appendix F). After patients left their appointment, the medical assistant
filled out a brief DNP project form and circled "Yes" or "No" questions to identify the
appropriate course of action and a multiple-choice question that delineated which tests patients

COLORECTAL CANCER SCREENING

15

chose most frequently (Appendix G). This CRC Screening Evaluation Form was used for dual
purpose, both for standard of care and for this project. No patient identification was present on
these Evaluation Forms.
The DNP student was present for the first 2 to 3 weeks of the practice change to answer
any questions or concerns the PCPs or medical assistants have. She was also available via email
and telephone. This process continued from October 2017 to February 2018. After project
completion, PCPs and medical assistants were given the PCA Survey again to evaluate for
personal perceptions on practice change.
Meaningful Use scores were provided to the DNP student by the office manager and IT
personnel to evaluate for any changes in scoring.
SPSS23 was used to store and analyze data. The DNP student analyzed data and ran
reports with the help of an assigned statistician at Arizona State University (ASU) College of
Nursing and Heath Innovation (CONHI). A paired t-test was used to measure pre and post survey
data. Means of quantitative data were reported from the CRC screening evaluation forms.
Meaningful Use scores pre and post project will be provided by the project site practice
administrator. After data analysis, an executive summary was provided to the primary care
practice to explain the results of the project and recommendations for further action. $396.36 was
spent on this project from the DNP student's personal funding. No outside funding was received
for this project.
Project Results
Meaningful Use
Meaningful Use scores were expected to improve from October to February, which is
when data collection and project implementation took place. Meaningful Use Scores prior to

COLORECTAL CANCER SCREENING

16

project implementation were provided to act as a baseline. 1/4/17 to 4/3/17 at 14.9%, 4/4/17 to
7/2/17 at 15.17%, and 7/3/17 to 9/30/17 at 15.58%. The first Meaningful Use scores that were
reported during the practice change were from 10/1/17 to 12/29/17 at 15.36%. Scores decreased
slightly from October to December, possibly due to the holiday season and increased incidence
of acute illness such as flu. Data from January and February 2018 were not able to be obtained
due to the practice only receiving reports biannually. Unfortunately, this was not made clear to
the DNP student early on so data from these two months will not be obtained. It was hoped that
data would have been highest in January and February 2018. Since this information was not
provided in time, the practice manager will evaluate 2018 data on her own and provide feedback
to the practice on her own terms.
CRC Screening Evaluation Forms
A total of 91 patients were screened during the project, N=91. Of those screened, 8% did
not meet screening criteria. 67% reported they had already had screening, but documentation was
not entered in the clinical decision support system (CDSS) section of the chart; documentation
was added by the MAs. This shows that the MAs have the ability to improve Meaningful use
scores for the practice simply by following this new workflow. If patients have been screened,
but it is not documented in the correct spot in the EHR, data is being missed and Meaningful Use
scores cannot improve. Adding documentation to CDSS ensures that Meaningful Use scores are
captured when reports are run and ensures that CRC screening status is being charted
consistently in the same spot by all providers. 70% of those screened were given a CRC
Screening Brochure by the MAs. The most common method of screening chosen was
colonoscopy, 87%. 3% chose FIT, 3% chose stool DNA, and 7% did not choose any form of
screening. A strong correlation was found between receiving a CRC brochure and choosing a

COLORECTAL CANCER SCREENING

17

form of screening. A Pearson correlation coefficient was calculated for the relationship between
patient's receiving a brochure and choosing a form of screening. A strong correlation was found
(r (-.293) = .01, p <.05) between receiving a CRC brochure and choosing a form of screening.
This indicates patients are more likely to choose a screening method when they receive a CRC
screening brochure.
Pre and Post PCA Surveys
The sample consisted of N=9; 4 (44%) PCPs and 5 (56%) MA's. One survey had to be
thrown out due to lack of a corresponding pre-survey. A paired-samples t test was calculated to
compare the mean pre-survey scores to mean post-survey scores. No significant difference from
pre to post survey satisfaction scores were found (t (8) = -1.542, p = .162). It was observed that
participants were reluctant to complete these surveys due to time and the number of questions
involved. Some participants took a few days to return surveys while one took two weeks to
return a post-survey. Two participants questioned the surveys, asking how they were relevant to
the project.
Discussion
This change in workflow allowed for staff and patient education regarding CRC.
Providers and MAs learned of new documentation techniques that would allow the eClinical
Works EHR to capture more data regarding Meaningful Use. While Meaningful Use scores are
not back yet for 2018, it is important for staff to document in the same spot, CDSS, for this data
to be captured. The practice manager reported that Meaningful Use data is only retrievable twice
a year, making it more difficult to monitor this information on a monthly basis. The project
champion and physician provider expressed liking the CRC Screening Brochures because they
were any easy way to approach the topic. The DNP student discussed with the practice manager

COLORECTAL CANCER SCREENING

18

if or where they would like to continue having the brochures available for patients. Patients and
providers overwhelmingly chose colonoscopy as the most common method of screening, which
research shows as the most common method currently. With the advancement of stool DNA
testing, these numbers may change in the future.
Originally the goal was to screen 300 patients at the clinic, however MAs began
screening patients less as time went on. When asked how the change in work flow was going,
simple answers were given such as "fine" or "most people are screened already". This may have
indicated a lack of understanding on the importance of continued screening and the need to raise
the topic of CRC screening. Data does not support that most people are screened in this office in
spite of the MAs thinking this is the case. With fewer CRC Screening Forms being returned as
time progressed, the decision was made to stop the project October with N=91 of patients
screened. MAs indicated this project did create extra work for them during their already busy
schedule and could be another factor for not reaching the original goal of 300 patients. In
reviewing the practice change, the MAs said filling out additional forms to capture data was time
consuming for them, and some expressed they "asked but forgot to fill out forms". Time is a
constant issue when making any workplace change. If this practice change were to continue,
forms and additional paperwork would be removed from the daily tasks. Instead, Meaningful
Use data would be the only method to capture uptake of screening patients and documenting in
the correct area of CDSS within eClinicalWorks. Many corporate organizations and companies
have mandatory compliance with Meaningful Use documentation and is stated so in their
organization policy. The fact that this practice is a private organization with no hospital
affiliation or larger organization overseeing their documentation process could be another reason
for decreased uptake of documentation. If the practice were to make this practice change an

COLORECTAL CANCER SCREENING

19

actual policy in their practice, uptake and compliance may increase, making this a sustainable
practice change.
Participants had trouble remembering their unique subject ID assigned to them to match
pre and post survey data. In the future a simpler ID code should be used to avoid confusion to
participants. There was no significance found between pre and post survey data. This could be
due to the length of time the providers and MAs have been at the clinic, and their perceived
belief that their work flow did not need to change, which is stated in the Ottawa Model of
Research Use (Appendix C). Follow up meetings and discussions regarding the importance of
capturing Meaningful Use data, how it affects patients and the practice as a whole, could help to
facilitate workplace change in the future. If this project were to be repeated a different pre and
post survey tool that was shorter in length would have been chosen. A post-survey asking about
perceptions of this project change with an open area for comments would have provided useful
feedback to see what participants liked or did not like about this project.
Conclusion
Further follow up is needed to evaluate cost effectiveness and sustainability of this
practice change. Data for Meaningful Use is only available biannually and must be requested
either by IT or the practice manager. The practice manger mentioned looking into this and
working with the newly hired IT staff to see if data can be reported more frequently. This would
allow for closer monitoring of this data and provide the practice with a more real-time accurate
result of CRC screenings on a monthly basis. The eClinical Works EHR does not automatically
report such data, so additional staff resources are required to continue measuring outcomes. It is
recommended that IT continue to pull this data, so the practice can have an accurate idea of
where they stand in capturing CRC screening Meaningful Use rates. Continued staff education is

COLORECTAL CANCER SCREENING

20

recommended to ensure everyone is documenting CRC screening status is the same location in
the EHR. Printing of CRC Screening Brochures would need to be calculated into annual budget
if the practice wishes to continue making them available for patients. Overall, 91 patients had the
topic of CRC screening brought to their attention, which is a great start in the early detection of
colorectal cancer. Only five providers in this practice chose to participate in this study. If all
providers in the practice implemented this change in work flow it is likely that more patients
would get screened on a more consistent basis. This practice is a private practice and does not
have to follow any specific policies such as larger organizations do. A set policy on CRC
screening and documentation could help to increase CRC screening rates at this clinic if the
practice chose to enact such a policy. Having a champion of change among the MAs in addition
to the PCPs could also have a positive result on sustaining this change. Ultimately the data shows
that receiving a CRC screening brochure likely increase the likelihood of a patient choosing a
form of screening, which was the goal of this project.

COLORECTAL CANCER SCREENING

21

References:

Administration on Aging. (2012). Arizona policy academy state profile. Retrieved from:
https://aoa.acl.gov/AoA_Programs/HPW/Behavioral/docs2/Arizona.pdf
American Gastroenterological Association. (2017). New CDC Report Shows Colorectal Cancer
Tests are Underused: 23 Million American Adults Have Never Been Tested. Retrieved
from: http://www.gastro.org/press_releases/new-cdc-report-shows-colorectal-cancertests-are-underused-23-million-american-adults-have-never-been-tested
Basch, C., Zybert, P., Wolf, R., Basch, C., Ullman, R., Shmukler, C., & ... Shea, S. (2015). A
Randomized Trial to Compare Alternative Educational Interventions to Increase
Colorectal Cancer Screening in a Hard-to-Reach Urban Minority Population with Health
Insurance. Journal Of Community Health, 40(5), 975-983. doi:10.1007/s10900-0150021-5
Baxter, N., Sutradhar, R., Li, Q., Daly, C., Honein-AbouHaidar, G., Richardson, D.,
&...Rabeneck, L. (2017). Do primary care provider strategies improve patient
participation in colorectal cancer screening? The American Journal of Gastroenterology.
doi:10.1038/ajg.2017.4

Dickinson, L. M., Dickinson, W. P., Nutting, P. A., Fisher, L., Harbrecht, M., Crabtree, B. F., …
West, D. R. (2015). Practice Context Affects Efforts to Improve Diabetes Care for
Primary Care Patients: A Pragmatic Cluster Randomized Trial. Journal of General
Internal Medicine, 30(4), 476–482. http://doi.org/10.1007/s11606-014-3131-3

COLORECTAL CANCER SCREENING

22

Fitzpatrick-Lewis, D., Ali, M. U., Warren, R., Kenny, M., Sherifali, D., & Raina, P. (2016).
Screening for Colorectal Cancer: A Systematic Review and Meta-Analysis. Clinical
Colorectal Cancer, 15(4), 298-313. doi:10.1016/j.clcc.2016.03.003
Gupta, S., Halm, E. A., Rockey, D. C., Hammons, M., Koch, M., Carter, E., & ... Skinner, C. S.
(2013). Comparative effectiveness of fecal immunochemical test outreach, colonoscopy
outreach, and usual care for boosting colorectal cancer screening among the underserved:
a randomized clinical trial. JAMA Internal Medicine, 173(18), 1725-1732.
Healthypeople.gov (2017). Retrieved from: https://www.healthypeople.gov/2020/topicsobjectives/topic/cancer
Hite, N. H., & Margolin, D. A. (2016). Advances in colonoscopy and screening for colon cancer.
Seminars in Colon & Rectal Surgery, 27(4), 181-186. doi:10.1053/j.scrs.2016.04.015
Hoffman, R., Steel, S., Yee, E., Massie, L., Schrader, R., & Murata, G. (2010). Colorectal cancer
screening adherence is higher with fecal immunochemical tests than guaiac-based fecal
occult blood tests: a randomized, controlled trial. Preventive Medicine, 50(5/6), 297-299.
doi:10.1016/j.ypmed.2010.03.010
Imperiale, T. F., Ransohoff, D. F., Itzkowitz, S. H., Levin, T. R., Lavin, P., Lidgard, G. P., . . .
Berger, B. M. (2014). Multitarget stool DNA testing for colorectal-cancer screening. New
England Journal of Medicine, 370(14), 1287-1297. doi:10.1056/NEJMoa1311194

Jaén, C., Crabtree, B., Palmer, R., Ferrer, R., Nutting, P., Miller, W., . . . Stange, K. (2010).
Methods for evaluating practice change toward a patient-centered medical home. Annals
of Family Medicine, 8 Suppl 1, S9-20.

COLORECTAL CANCER SCREENING

23

Logan, J., Harrison, M., Graham, I., Dunn, K., & Bissonnette, J., (1999). Evidence-based
pressure-ulcer practice: the ottawa model of research use. Canadian Journal of Nursing
Research, 31(1), 37-52.
Malik, P. (2016). A novel multitarget stool DNA test for colorectal cancer screening.
Postgraduate Medicine, 128(2), 268-272. doi:10.1080/00325481.2016.1135035
Patel, S. S., & Kilgore, M. L. (2015). Cost effectiveness of colorectal cancer screening strategies.
Cancer Control: Journal of the Moffitt Cancer Center, 22(2), 248-258.
Potter, M. B., Phengrasamy, L., Hudes, E. S., McPhee, S. J., & Walsh, J. M. (2009). Offering
annual fecal occult blood tests at annual flu shot clinics increases colorectal cancer
screening rates. Annals of Family Medicine, 7(1), 17-23. doi:10.1370/afm.934
Ramdass, P., Petraro, P., Via, C., Shahrokni, A., & Nawaz, H. (2014). Providers Role in
Colonoscopy Screening for Colorectal Cancer. American Journal Of Health Behavior,
38(2), 234-244. doi:10.5993/AJHB.38.2.9
Schroy, P. C., Duhovic, E., Chen, C. A., Heeren, T. C., Lopez, W., Apodaca, D. L., & Wong, J.
B. (2016). Risk Stratification and Shared Decision Making for Colorectal Cancer Screening.
Medical Decision Making, 36(4), 526-535. doi:10.1177/0272989X15625622
Shaw, E. K., Ohman-Strickland, P., Piasecki, A., Hudson, S. V., Ferrante, J. M., McDaniel Jr,
R.,R., . . . McDaniel, R. R.,Jr. (2013). Effects of facilitated team meetings and learning
collaboratives on colorectal cancer screening rates in primary care practices: A cluster
randomized trial. Annals of Family Medicine, 11(3), 220-228. doi:10.1370/afm.1505
Singal, A. G., Gupta, S., Tiro, J. A., Skinner, C. S., McCallister, K., Sanders, J. M., & ... Halm,
E. A. (2016). Outreach invitations for FIT and colonoscopy improve colorectal cancer

COLORECTAL CANCER SCREENING

24

screening rates: A randomized controlled trial in a safety-net health system. Cancer
(0008543X), 122(3), 456-463. doi:10.1002/cncr.29770
Spruce, L. R., & Sanford, J. T. (2012). An intervention to change the approach to colorectal
cancer screening in primary care. Journal of the American Academy of Nurse Practitioners,
24(4), 167-174. doi:10.1111/j.1745-7599.2012.00714.x
Stacey, D., Pomey, M., O'Connor, A. M., & Graham, I. D. (2006). Adoption and sustainability of
decision support for patients facing health decisions: An implementation case study in
nursing. Implementation Science: IS, 1(1), 17-17. doi:10.1186/1748-5908-1-17
Symonds, E. L. E. (2015). Asian pacific journal of cancer prevention: APJCP: Improving
participation in colorectal cancer screening: A randomised controlled trial of sequential
offers of faecal then blood based non-invasive tests. Asian Pacific Organization for Cancer
Prevention.
Ueland, A., Hornung, P., & Greenwald, B. (2006). Colorectal cancer prevention and screening: a
health belief model-based research study to increase disease awareness. Gastroenterology
Nursing, 29(5), 357-363.
Urban Adolescent SRH SBCC Implementation Kit. (2016). Health belief model. Retrieved from:
https://sbccimplementationkits.org/urban-youth/health-belief-model/
US Preventive Services Task Force, F., Bibbins-Domingo, K., Grossman, D. C., Curry, S. J.,
Davidson, K. W., Epling, J. W.,Jr, . . . Siu, A. L. (2016). Screening for colorectal cancer: US
preventive services task force recommendation statement. JAMA: Journal of the American
Medical Association, 315(23), 2564-2575. doi:10.1001/jama.2016.5989
Volk, R. J., Linder, S. K., Lopez-Olivo, M., Kamath, G. R., Reuland, D. S., Saraykar, S. S., . . .
Pignone, M. P. (2016). Patient decision aids for colorectal cancer screening: A systematic

COLORECTAL CANCER SCREENING

25

review and meta-analysis. American Journal of Preventive Medicine, 51(5), 779-791.
doi:10.1016/j.amepre.2016.06.022
World Cancer Research Fund International. (2017). Colorectal cancer statistics. Retrieved from:
http://www.wcrf.org/int/cancer-facts-figures/data-specific-cancers/colorectal-cancerstatistics

COLORECTAL CANCER SCREENING

26

Appendix B
Synthesis Table

COLORECTAL CANCER SCREENING

27

Appendix C
Health Belief Model

COLORECTAL CANCER SCREENING

28

Appendix D
Ottawa Model of Research Use

COLORECTAL CANCER SCREENING

29

Appendix E
PCA Survey

COLORECTAL CANCER SCREENING

30

Appendix F
CRC Screening Brochure

COLORECTAL CANCER SCREENING

31

Appendix G
CRC Screening Evaluation Form

COLORECTAL CANCER SCREENING

32

Appendix H
MA CRC Screening Flowsheet

Citation

Theory/ Design/
Concep Method
tual
Frame
work

Basch et al., (2015).
A Randomized trial
to compare
alternative
educational
interventions to
increase colorectal
cancer screening in a
hard-to-reach urban
minority population
with health insurance.
Funding: American
Cancer Society
Grant: Identifier
NCT02392143
Bias/Conflict: $20
promised to PTs to
complete study
USA- NYC

Health
Promotio
n ModelNola J.
Pender

Citation

Theory/ Design/

Design: RCT
Purpose: To
compare
interventions to
increase CRC
screenings in
those with health
insurance

Appendix A
Evaluation Table: Colorectal Cancer Screening
Sample/ Setting
Major
Measurement Data
Variables &
/
Analysis
Definitions
Instrumentati
on

Findings/
Results

Level of
Evidence;
Decision for
practice/
application to
practice

N= 8792
n=564 PTs randomized
Inclusion criteria: Adults age
50-75 years with health
insurance (union-based selfadministered and self-insured
benefit fund). Currently out of
compliance with recommended
CRC screening. Having a
"regular doctor", a stated
intention to remain in the
benefit fund for at least 1 year.
Reachable by telephone, able to
communicate in English, ability
to give consent.
Exclusion criteria: No health
insurance. Hx of colorectal
polyps, IBD, IBS, Crohn's
disease, UC, or current
treatment for any type of CA.

IV 1: PEM1
Only
IV 2:AD Only
IV 3:AD &
TTE

N%
screened for
CRC:
PEM only
N=18.3%.
AD only
N=20%
AD&TTE
N=25.6%.
Total N=
21.5%
No
statistically
significant
pairwise
differences
between
groups in
screening
rate

Level I
Strengths: RCT.
TTE based on
previously tested
model.
Weaknesses: No
theoretical
framework named.
Poor follow up with
participants after 1
year. NYC already
has high rate of
CRC screening in
the country.
Conclusion:
Screening rate was
almost 40% higher
in AD&TTE group
vs PEM- but was
not SS

Sample/ Setting

Major

Findings/

Level of

DV: CRC
screening status

3 types data
collection:
Baseline survey
data.
Implementation
data.
Outcome data
based on
medical claims.

Pairwise group
differences
using 2x2 χ2
analyses.
Linear trends
across groups.

*PEM1highlighted
colonoscopy as
being the only
test that can
identify &
prevent CRC,
but explained
other CRC
screening tests

Measurement Data

ACN-advanced colorectal neoplasia, ACNI-Advanced colorectal neoplasia index, AD- academic detailing, BE- barium enema, CA- cancer, CAPE-Client
Agency Program Enrollment, CI- confidence interval, CRC-Colorectal cancer, CS-colonoscopy, CT- computed tomography, DA-decision aid, DNAdeoxyribonucleic acid, DRE- digital rectal exam, EMR- electronic medical record, FAP- familial adenomatous polyposis, FIT- fecal immunochemical test,
FOBT-fecal occult blood test, FS- flexible sigmoidoscopy, gFOBT, guaiac fecal occult blood testing, HA- hazard ratio, HNPCC- hereditary nonpolyposis
colorectal cancer, IBD-inflammatory bowel disease, IBS- irritable bowel syndrome, ICES-Institute for Clinical Evaluation Sciences, iFOBT- immunochemical
fecal occult blood test, IPDB-ICES Physicians’ Database, N-number of participants, n-subset of participants, NNS- number needed to screen, OR-odds ratio,
OHIP- Ontario Health Insurance Plan, PCP- primary care physician, PEM1- printed educational materials, PEM2- Patient Enrollment Model, PT-patient,
QBE- question-behavior effect, QI-quality improvement, RCT-randomized control trial, RPDB- Registered Persons Database, SCOPE-The Supporting
Colorectal Cancer Outcomes through Participatory Enhancements, SDM-shared decision making, SDMP-satisfaction with decision making process, SSstatistically significant, TTE-tailored telephone education, UC- ulcerative colitis, UTD-up to date

COLORECTAL CANCER SCREENING

Baxter et al., (2017).
Do primary care
provider strategies
improve patient
participation in
colorectal cancer
screening?
Funding: Canadian
Cancer Society (grant
#2011-700803).
Cancer Care Ontario
& Canadian Institutes
of Health Research
Foundation Grant
(Competition
#201509).
Bias/Conflict: None
Setting: Ontario,
Canada

Concep
tual
Frame
work

Method

Adaptati
on
ModelRoy

Design: Crosssectional survey

Purpose: To
determine the
effect of provider
strategies to
increase screening
in a single-payer
system.

n=717 PCPs
n= 147,834 rostered PTs
Inclusion Criteria:
PCPs: practicing PCPs with a
valid unique physician
identifier in IPDB. Belong to a
PEM2 identified in CAPE.
PTs: Receiving comprehensive
primary care including CRC
from a specific PCP. Eligible
for CRC screenings (age 50-74)
at any point from 4/1/2012 to
3/31/2013 according to RPDB.
Exclusion Criteria:
PCPs: retired, not complete
initial survey, not in Ontario
province, main practice not
primary care.
PTs: Residence not Ontario as
of 3/31/2013, no contact with
healthcare system within > 6
years prior to 3/21/13, if first
OHIP eligibility was later than
3/31/2013, if dx of invasive
CRC prior to 4/1/2012.
Underwent colonic resection
prior to 4/1/2012l, if PT died
before 3/31/2013.

34

Variables &
Definitions

/
Instrumentati
on

Analysis

Results

Evidence;
Decision for
practice/
application to
practice

IV1: Survey to
PCP to identify
covariates

Mailed surveys.

SAS statistical
software.

Multiple
PCP
strategies
(HR=1.27,
95%
CI:1.161.39,
P<0.0001
for PCPs
using 4-5 vs
0-1
strategies).
Systematic
approach for
screening
weakly
associated
with
screening
uptake
(HR=1.14,
95%CI:
1.03-1.26,
P=0.04>5
years
overdue vs
<1 year
overdue

Level VI
Strengths: Large
study. Strong
heterogeneity of PT
population.
Weaknesses: No
theoretical
framework used.
Population based
survey. Different
types of CRC
screenings not
explained.
Conclusion: No
single strategy was
strongly associated
with screening.
PCP use of multiple
strategies >
screening uptake.

DV: Time to
become up to
date with CRC
screening status

Modified
Dillman-style
multimodal
approach.
Cox
proportional
hazard models.

Descriptive
Statistics.
Regression
analyses

COLORECTAL CANCER SCREENING

35

Citation

Theory/ Design/
Concep Method
tual
Frame
work

Sample/ Setting

Major
Variables &
Definitions

Measurement Data
/
Analysis
Instrumentati
on

Findings/
Results

Level of
Evidence;
Decision for
practice/
application to
practice

Fitzpatrcik-Lewis et
al., (2016). Screening
for colorectal cancer:
a systematic review
and meta-analysis

System's
ModelNeuman

87 Included Studies

IV1:
effectiveness of
each CRC
screening test to
reduce CRCspecific
mortality, allcause mortality,
or incidence of
late-stage CRC
IV2: Optimal
age to start &
stop screening
IV3: Optimal
screening
interval
IV4: Benefits
of different
screening tests
IV5: Incidence
of harm
DV: CRC
screening
recommendatio
ns

Data extraction
completed by
one reviewer
and verified by
a second using
standardized
forms- all
checked by
statistician
before analysis.

IV1: gFOBT
& FS 18%
reduction
IV2: older
adults ≥ 60
IV3: No
change
between
biennial and
annual
gFOBT
screening
IV4: gFOBT
& FS most
benefit
IV5: FS
potential
harm
DV: iFOBT
higher
sensitivity
and
specificity

Level V
Strengths: Large
review.
Weaknesses: No
consideration of PT
preference. No
consideration of
implementation.
Conclusion: CRC
screening using
FOBT and FS is
effective for
reducing CRC
mortality and
incidence of latestage disease

Funding: Public
Health Agency of
Canada and the
Canadian Institutes of
Health Research.
Bias/Conflict: None.
Setting: Ontario,
Canada

Design:
Systematic review
and meta-analysis

Purpose: To
evaluate the
effectiveness of
CRC screening in
asymptomatic
adults.

Inclusion criteria: Average
risk, asymptomatic adults age >
18 years. CRC screening with a
CS, CT, colonography, gFOBT,
iFOBT, FS, BE, DRE, fecal
DNA & other identified tests.
Primary care setting a PCP
could refer to for colonoscopy,
CT and FS testing. RCT with
comparison groups of no
screening or comparison
between tests. Harms: any
study.
Exclusion criteria: High risk
adults with FAP, HNPCC,
history of IBD, personal history
of polyps or CRC; adults with
symptoms suggesting
underlying CRC or with known
genetic mutations associated
with increased CRC risk. Casefinding or surveillance tests.

RCTs appraised
using Cochrane
Risk of Bias
tool & Grade
Recommendatio
ns Assessment,
Development
and Evaluation
(GRADE)
system.

Risk ratios
using
DerSimonian
and Laird
random-effects
model with
inverse
variance
method.
Binomial CIs
calculated by
Wilson score
interval
method.
Review
Manager 5.3,
STATA 12,
and
GRADEpro

COLORECTAL CANCER SCREENING

36

Citation

Theory/ Design/
Concep Method
tual
Frame
work

Sample/ Setting

Major
Variables &
Definitions

Measurement Data
/
Analysis
Instrumentati
on

Findings/
Results

Level of
Evidence;
Decision for
practice/
application to
practice

Gupta et al., (2013).
Comparative
effectiveness of fecal
immunochemical test
outreach,
colonoscopy
outreach, and usual
care for boosting
colorectal cancer
screening among the
underserved
Funding: Cancer
Prevention and
Research Institute of
Texas, National
Institutes of HealthGrants

Health
promotio
n modelPender.

Design: RCT

N=5994

SAS PROC
PLAN

Purpose: Increase
CRC screeningthrough mailed
outreach & FIT
offering vs
colonoscopy

Inclusion Criteria: uninsured
PTs not UTD with CRC
screening, age 54-64 years,
served by the John Pete Smith
Health Network, a safety net
health system.

IV1: FIT
outreach
IV2: CS
outreach
IV3: Usual care
DV: CRC
screening test
within 1 year

FIT tripled
CRC
screening
rates, CS
doubled
rates
compared
with usual
care (40.7%,
24.1%, and
12.1%).

Level I
Strengths: Size of
study. Level of
evidence.
Weakness: No
theoretical
framework noted.
Less PTs were
randomized to CS
outreach overall vs
other groups.
Uninsured PTs
unlikely to get
colonoscopy
regardless due to
lack of payment
method.
Conclusion: FIT >
vs CS. No

Bias/Conflict: None

Exclusion Criteria: UTD CRC
screening status, no address or
phone number on file, language
other than English or Spanish,
history of CRC, IBD, polyps,
no recent health system visit,
incarceration.

SAS Software

CPT codes

2-sample t test
χ2 or Fisher
exact test

FIT test
interpretation
by healthy
system clinic
laboratory

2 primary
comparisons,
P<.025 SS
All other
comparisons,
P<.05 SS

P<.001

COLORECTAL CANCER SCREENING

37

intervention results
in overall less CRC
screening

Setting: USA- Texas

Citation

Theory/ Design/
Concep Method
tual
Frame
work

Sample/ Setting

Major
Variables &
Definitions

Measurement Data
/
Analysis
Instrumentati
on

Findings/
Results

Level of
Evidence;
Decision for
practice/
application to
practice

Hoffman et al.,
(2010). Comparative
Effectiveness of
Multifaceted
Outreach to Initiate
Colorectal Cancer
Screening in
Community Health
Centers.
Bias/Conflict: None
Funding:
Department of
Veterans Affairs

Health
promotio
n modelPender.

FIT n= 202
gFOBT n= 202

IV1: FIT
IV2: gFOBT
DV: CRC
screening
adherence

Random digit
number
generator
(http://www.ran
domizer.org/)
used to assign
subjects to
receive either
FIT or gFOBT.

Screening
adherence
was higher
with FIT
than gFOBT
(61.4% vs.
50.5%, P =
0.03). The
adjusted
odds ratio
for
completing
FIT vs.
gFOBT was
1.56, 95%
CI 1.04,
2.32.

Level I
Strengths: Level of
evidence, size of
study.
Weaknesses: VA
population only.
Motivated PTs only
consented to study.
Conclusion: FIT >
gFOBT CRC
screening status.

Findings/
Results

Level of
Evidence;
Decision for

Design: RCT
Purpose:
Determine
whether CRC
screening
adherence is
greater with FIT
or gFOBT

(project number
SHP 08-177)

Setting: New Mexico Veterans
Affairs Health Care System
Primary Care Clinic
Inclusion: ICD-9 problem lists,
CPT procedure codes,
pathology records, and
screening reminders to identify
primary care PTs age 50-80
years due for CRC screening.
Exclusion: PTs with CRC,
IBD, polyps.

FIT tests
analyzed on the
OC-Auto Micro
80 instrument;
human
hemoglobin
levels N 100
ng/ml were
considered +.

USA

Citation

Theory/ Design/
Concep Method
tual

Sample/ Setting

Major
Variables &
Definitions

-Descriptive
statistics with
95% CI.
-t-tests or
Wilcoxon twosample test for
continuous
variables, &
chi- square
tests for
categorical
variables.

-Multivariate
logistic
regression
modeling to
compare
adherence

Measurement Data
/
Analysis
Instrumentati

COLORECTAL CANCER SCREENING

38

Frame
work
Imperiale et al.,
(2014). Multitarget
stool DNA testing for
colorectal-cancer
screening
Funding: Exact
Sciences;
ClinicalTrials.gov
number,
NCT01397747

Health
promotio
n modelPender.

on

Design: Crosssectional
Purpose: To
evaluate the
multitarget stool
DNA test as a tool
for screening for
CRC

Bias/Conflict:
Funded by Exact
Sciences- maker of
the stool DNA test
(Cologuard- though
not named in study
directly)

Citation

Theory/ Design/
Concep Method
tual
Frame
work

practice/
application to
practice

N= 9989
Setting: 90 sites in US &
Canada
Inclusion Criteria:
Asymptomatic PTs ages 50-84
years considered average risk
for CRC & who were scheduled
to undergo screening CS.
Exclusion Criteria: History of
colorectal neoplasia, digestive
CA, IBD; had undergone CS
within previous 9 years or
barium enema, FS within 5
years; positive results of fecal
blood testing within 6 months;
undergone colorectal resection
for any reason other than
sigmoid diverticula; rectal
bleeding within 30 days;
personal or family history of
CRC; had participated in any
interventional clinical study
within 30 days; or unable or
unwilling to provide written
informed consent.

IV1:
Multitarget
DNA test
IV2: FIT
DV1:
Sensitivity for
detecting CRC
DV2:
Sensitivity for
detecting
advanced
precancerous
lesions
DV3: Rate of
detection of
polyps

One-sded
McNemar
pariedcomparions
tests.

SAS &
StatXact
software.

Sample/ Setting

Major
Variables &
Definitions

Measurement Data
/
Analysis
Instrumentati
on

HanleyMcNeils
method to
calcuate P
values

IV1:
Identified 60
of 65 PTs
with CRC;
95% CI
IV2: FIT
identified 48
of 65 CRC;
95%CI
DV1: 92.3%
DNA,
73.8%
FIT(P=0.002
).
DV2: 42/4%
DNA;
23.8% FIT
(P=<0.001)
DV3:
69.2%
DNA;
46.2% FIT
(P=0.004)
NNS: 154
with CS;
166 DNA;
208 FIT.

Level IV
Strengths: Use of
additional CRC
screening tool now
available & shows
promise.
Weaknesses: Has
bias. Does not
discuss
implementation to
increase CRC
screening rates.
Does not discuss PT
preferences. What is
the cost?
Conclusion: Stool
DNA test has higher
sensitivity but lower
specificity for CRC
detection.

Findings/
Results

Level of
Evidence;
Decision for
practice/
application to
practice

COLORECTAL CANCER SCREENING

Potter et al., (2009).
Offering annual fecal
occult blood tests at
annual flu shot clinics
increases CRC
screening rates
Funding: Grants
from the American
Cancer society,
Cancer Control
Career Development
Award for Primary
Care Physicians, and
the National Cancer
Institute, Asian
American Network
for Cancer
Awareness, Research
& Training.
Bias/Conflict: None.

General
Model of
the
Determin
ants of
Behavior
al
Change
(a
synthesis
of
behavior
al
theories)

Design: Time
RCT
Purpose: To
determine whether
providing home
FOBT kits to
eligible PTs
during influenza
inoculation clinics
can contribute to
higher CRC
screening rates.

N= 514
Control n= 246
Intervention n= 268
Setting: Flu shot clinic
Inclusion Criteria: PTs age
50-79 years who were mailed
flu shot campaign
announcements & who attended
the 17 flu shot clinic sessions.
Had not had a FOBT since the
end of prior flu season, a
colonoscopy in past 10 years,
any previously unevaluated
abnormal FOBT results, or
history of recent unevaluated
rectal bleeding.
Exclusion Criteria: Did not
attend one of the 17 flu shot
clinic sessions.

39

IV1: PTs
offered flu shots
IV2: PTs
offered flu shots
and FOBT kits
DV: CRC
screening status

Flu shot
campaign
announcements.
Staff training
sessions.
Colorful
multilingual
education
sheets.

Stata software.
Pearson χ2 for
categorical
variables.
2-sample
Wilcoxon test
and McNemar
test.

FOBT kits with
postage-paid
return
envelopes.

Total PTs
initially not
UTD;
Control n=
20.7 % PTs
who become
UTD
Intervention
n= 68% PTs
who became
UTD P
<.001
Total PTs
initially
UTD;
Control
n=90.0%,
Intervention
n=98%,
P<.005

Level I
Strengths: Strong
theoretical
framework. Well
explained and
deigned study.
Weaknesses:
Limited populationonly those who
attend flu clinicsmissing large
portion of
population.
Conclusion: FOBT
kits during flu
season increased
CRC rate. Needs to
reach a broader
population however.
Multidisciplinary
team approach
works.

USA

Citation

Theory/ Design/
Concep Method
tual
Frame
work

Sample/ Setting

Major
Variables &
Definitions

Measurement Data
/
Analysis
Instrumentati
on

Findings/
Results

Level of
Evidence;
Decision for
practice/
application to
practice

Schroy et al., (2016).
Risk stratification
and shared decision
making for CRC
screening: a
randomized
controlled trial

Health
promotio
n modelPender.

n=352
Setting: Boston Medical
Center-private, not-for-profit
community-based academic
medical center. Largest safety
net hospital in New England.
Inclusion criteria:
Asymptomatic average-risk

IV1: Decision
aide alone
IV2: Decision
aide plus risk
assessment tool
DV1:
Concordance
between PT

Electronic risk
assessment
tool- pre and
post-test for
PTs.
Web-based
DVD-formatted
decision aid for

IV1:
Concordanc
e 88%
IV2:
Concordanc
e 85.6%
DV1: DA
only 88.0;

Level I

Design:
RCT, unblinded,
parallel-group
Purpose: To
determine whether
risk stratification
for ACN

SAS software.
P < 0.05 level

Strengths: Strong
level of evidence.
Validated tools
used.
Weaknesses:
Unblinded. Single

COLORECTAL CANCER SCREENING

Funding: National
Cancer Institute
Grant (RO1
CA131197).
Bias/Conflict: None
USA

influences
provider
willingness to
comply with PT
preferences when
selecting a desired
CRC screening
option

primary care PTs cared for at
Boston Medical Center. Age
50-75 years, under care of PCP,
and due for CRC screening.
Exclusion criteria: Lower GI
symptoms or iron deficiency
anemia, increased risk of CRC
for which CS was the preferred
screening method, lack of
fluency in written & spoken
English, comorbidities that
precluded CRC by any method

40

preference &
test ordered
DV2: PT
satisfaction
with decisionmaking process.
DV3: Screening
intentions
DV4: Test
completion
rates
DV5: Provider
satisfaction
Definition:
Concordance is
defined as the
number of
patients who
had their
preferred test
ordered (n)
divided by the
total number of
patients who
preferred the
test (N).

physiciansWhat Colon
Cancer
Screening Test
is Best for You an educational
website of the
Section of
Gastroenterolog
y at Boston
University
School of
Medicine and
Boston Medical
Center that was
created with
funding for the
Agency for
Healthcare
Quality and
Research.
ACNI P=0.720.93
Absolute risks
assessment
scores
Brief paperbased 3-item
survey for
providers
12-item SDMP
scale

DA + ACNI
85.6 P=0.40.
DA + ACNI
High risk
83.5, DA +
ACNI Low
Risk 87.1
P=0.51
DV2:
Concordanc
e n= 52.0,
Discordance
n=48.9,
P<0.001
DV3:
Concordanc
e n= 4.6,
discordance
n=4.0,
P<0.001
DV4:
Concordanc
e n= 109,
discordance
n=7,
P=0.004
DV5: SDM
useful
P=0.70,
SDM will
reduce time
to decide on
CRC
screening
P=0.10,
reduces
malpractice
risk posttest

institution setting.
Lack of provider
education for the
study. Decision aid
tool took between
11-34 minutes- too
much time in busy
primary care setting
Conclusion: PT
preference an
important factor to
increase CRC rates.
Web-DVD gave PT
ALL the options for
CRC screeninggreat web-based tool
for education-this
gave PTs multiple
options for CRC
screening

COLORECTAL CANCER SCREENING

41

(validated)

n=3.2.
SDM
positively
effects CRC
screening
rates.

Citation

Theory/ Design/
Concep Method
tual
Frame
work

Sample/ Setting

Major
Variables &
Definitions

Measurement Data
/
Analysis
Instrumentati
on

Findings/
Results

Level of
Evidence;
Decision for
practice/
application to
practice

Singal et al., (2016).
Outreach invitations
for FIT and
Colonoscopy
improve colorectal
cancer screening
rates.
Funding: National
Cancer Institute
National Institutes of
Health grant.
Polymedco Corp.
provided FIT test kits
Bias/Conflict: 1
author a paid member
of Scientific
Advisory Board for
Exact Sciences for
work performed
outside current study
USA

Health
promotio
n modelPender.

N=5999
Setting: Primary care &
hospital
Inclusion Criteria: PTs age
50-64 years, with at least 1 visit
to PCP clinic within the year
before randomization, residents
of Dallas County, Parkland
Health Plus coverage.
Exclusion Criteria: UPD with
CRC screening, no address or
telephone number on file,
language other than English or
Spanish, history of CRC, IBD,
polyps, or prior colectomy,
incarceration.

IV1: Usual care
IV2: FIT
outreach
IV3: CS
outreach

Querying
electronic
healthy system
laboratory data
for FIT testing
& combination
of test orders &
administrative
claims data for
FS or CS

IV1:
Screening
rates 29.6%
IV2:
Screening
rate 58.8%.
OR 3.39,
95% CI
IV3:
Screening
rate 42.4%.
OR 1.00,
95%CI

Level I
Strengths: 3-year
study. Well
designed study.

Design:
Prospective RCT
Purpose:
Compare initial
screening
participation
across the 3
groups among
individuals with at
least 1 year of
postintervention
follow-up

DV: CRC
screening status
within 12
months

Pearson chisquare test
SAS statistical
software
alpha .05

Weaknesses: Lack
of insurance could
have skewed CS
option
Conclusion: FIT
option increases
CRC screening
most, followed by
CS.

COLORECTAL CANCER SCREENING

42

Citation

Theory/ Design/
Sample/ Setting
Concep Method/Sampli
tual
ng
Frame
work

Major
Variables &
Definitions

Measurement Data
/
Analysis
Instrumentati
on

Findings/
Results

Level of
Evidence;
Decision for
practice/
application to
practice

Volk et al., (2016).
Patient decision aids
for colorectal cancer
screening: A
systematic review
and meta-analysis

Systems
Model of
Clinical
Preventiv
e Care

Studies
evaluating
patient decision
aids for CRC
screening in
average-risk
adults and their
impact on
knowledge,
screening
intentions, and
uptake.
IV1: Received
decision aid
IV2: Received
general CRC
screening
information
DV: CRC
screening rates

Variable per
study;
individual
measurement
tools not listed.

PT exposed
to DA =
greater
knowledge
(mean
difference =
18.3 of 100;
95%CI=15.5
, 21.1), more
likely to be
interested in
screening
(pooled
relative risk
= 1.5;
95%CI =1.2,
2.0), & more
likely to be
screening
(pooled
relative
risk=1.3;
95%
CI=1.1,1.4)

Level V
Strengths: Large
amount of studies
included. 7 CRC
screening options
included in studies.
Weaknesses:
Multiple types of
studies included.
Multiple outcome
scales used in
studies.
Conclusion:
Decision aids
associated with
greater intentions to
be screened &
screening uptake

Funding: Award
#R21CA132669 to
Dr. Robert J. Volk
form National Cancer
Institute.

Bias/Conflict: None.
USA

Design:
Systematic review
and meta-analysis

23 Articles; 21 trials including
11,900 subjects.

Purpose: To
evaluate the effect
of patient decision
aids on CRC
screening rates

Inclusion Criteria:
Quantitatively evaluated
decision aid comparted to one
or more conditions within a
pre-post evaluation.
Exclusion Criteria: Not
relevant, an intervention that
did not include trade-off
information, decision aides did
not have at least a pre-post
design, decision aids did not
report primary outcome
findings.

Meta-analysis
using a
standardized
form.