Journal of Urban Health: Bulletin of the New York Academy of Medicine, Vol. 82, No. 2,
The Author 2005. Published by Oxford University Press on behalf of the New York Academy of Medicine. All rights
reserved. For permissions, please e-mail: email@example.com
Advance Access publication May 11, 2005
Use of a Patient Navigator to Increase Colorectal
Cancer Screening in an Urban Neighborhood
Lina Jandorf, Yahaira Gutierrez, Jaime Lopez,
Jennifer Christie, and Steven H. Itzkowitz
ABSTRACT Colorectal cancer (CRC) is the second leading cause of cancer-related deaths
in the United States. Racial disparities in CRC incidence and mortality have been well
documented. In addition, lower rates of CRC screening among ethnic minorities have
been reported. Therefore, we tested the effectiveness of a patient navigator (PN) in
increasing compliance with CRC screening in a minority community health setting. Men
and women aged 50 or older attending a primary care practice were enrolled if they had
not had a fecal occult blood test within the past year, a sigmoidoscopy or barium enema
within the past 3–5 years, or a colonoscopy within the past 10 years. Participants were
randomly assigned either to receive navigator services (PN+) or not to receive navigator
services (PN−). There were no demographic differences between the two groups. Within
6 months of physician recommendation, 15.8% in the PN+ group had complied with an
endoscopic examination, compared with only 5% in the PN− group (P=.019). The
PN+ group also demonstrated higher rates of fecal occult blood test completion (42.1%
vs. 25%, P=.086). Thus, a PN system successfully increases CRC screening rates
among a predominantly minority population of low socioeconomic status.
KEYWORDS Colorectal cancer screening, Fecal occult blood test, Flexible sigmoidos-
copy, Urban health.
Colorectal cancer (CRC) is the second leading cause of cancer-related deaths in the
United States.1 It has been estimated that over 147,000 new cases of CRC would
occur in 2003, with over 57,000 deaths due to this disease.2 CRC screening tests,
which have now been demonstrated to save lives, are still, unfortunately, woefully
underutilized.3 Racial disparities in CRC incidence and mortality have been well
documented. CRC incidence and mortality is highest for African American men
and women compared with White, Hispanic, American Indian, and Asian individuals
in the United States (Surveillance, Epidemiology, and End Results 1973–1996;
http://seer.cancer.gov/csr/1973_1996/colorect.pdf). On the basis of 1997 Surveil-
lance, Epidemiology, and End Results data, Hispanic men and women have a
Ms. Jandorf is with the Department of Oncological Sciences, Mount Sinai School of Medicine, New York,
New York; Ms. Gutierrez, and Drs. Christie and Itzkowitz are with the Division of Gastroenterology,
Mount Sinai School of Medicine, New York, New York; Dr. Lopez is with Settlement Health Center,
New York, New York.
Correspondence: Lina Jandorf, MA, Department of Oncological Sciences, Mount Sinai School of
Medicine, One Gustave Levy Place, Box 1130, New York, NY 10029. (E-mail: firstname.lastname@example.org)
PATIENT NAVIGATOR USE IN COLORECTAL CANCER SCREENING 217
lower incidence of and mortality from CRC compared with Caucasians and
African Americans.4,5 However, because many Hispanic Americans are immigrants
from countries of lower CRC risk, it is anticipated that as they remain in the
United States, and with subsequent generations, their CRC risk will rise to the level
that is currently prevalent in the general US population.
Racial disparities also exist with respect to CRC screening. The disparity
between screening rates for African American and White individuals is in part
explained by the generally lower adherence to cancer screening tests among African
American men and women. According to the Behavioral Risk Factor Surveillance
System data from the National Center for Chronic Disease Prevention and Health
Promotion, only 25.9% of African Americans in the United States who are over the
age of 50 report ever having used a fecal occult blood test (FOBT) kit and 30%
have ever reported having a flexible sigmoidoscopy (FS) and/or colonoscopy.3 Even
lower rates are reported for Hispanic men and women over age 50; only 20% have
ever completed a FOBT home test kit and 25% have ever had an FS and/or colonos-
The lower rate of CRC screening among ethnic minorities is unfortunate given
the substantial evidence that regular screening can successfully reduce mortality
from CRC. Four prospective randomized controlled trials conducted in the United
States and in Europe have demonstrated that CRC mortality can be reduced by
approximately 33% if FOBT is performed annually and by 15–18% if it is per-
formed every other year.6–9 Although there are no prospective controlled studies of
screening FS to date, retrospective case-control studies indicate that sigmoidoscopy
can reduce deaths due to cancers within reach of the instrument by approximately
The East Harlem Partnership for Cancer Awareness was established to help
identify barriers to cancer screening in minority, medically underserved people in
the East Harlem community. Information obtained during five focus group sessions
with 43 African American and Hispanic men and women suggested that one possible
barrier to CRC screening in this population is the difficulty patients have trying to
navigate the health care system. Therefore, we reasoned that a patient navigator
(PN) system might enhance the completion of CRC screening tests.
A PN is an individual who can work individually with patients to both educate
and help them negotiate the medical system. PN programs have been shown to
increase screening rates for prostate and breast cancer. In one study of prostate cancer
screening, over 1,500 men were recruited from community sites (churches, barber-
shops, housing projects, etc.).12 Subjects were then divided into groups to receive
one of four educational interventions: traditional, peer educator, client navigator,
or combination. The traditional method involved educating patients about prostate
cancer. The peer educator intervention included the traditional method along with a
personal testimony concerning the importance of going for screening by a man who
could be considered a “peer” of the sample. The client navigator approach included
the traditional method along with a social worker who “navigated” the health care
system for the sample to participate in screening. The navigator provided written
reminders, telephone calls, and/or transportation to their subjects, encouraging par-
ticipation in prostate cancer screening. The combination method included the tradi-
tional, peer educator, and client navigator intervention. Of the four educational
interventions, two were significant predictors of participation in prostate cancer
screening: the client navigator (P=.000, OR=2.039) and the combination method
218 JANDORF ET AL.
Other investigators implemented a PN system in a medically underserved com-
munity in the Bronx, New York, to assist women in the follow up of significant
abnormalities on mammography.13 The population in this study was primarily
Black and Hispanic, uninsured, and of low educational status. The PN contacted all
women through telephone, ensured timely appointments, walked women through
diagnostic tests, ensured follow up on all referrals, and checked to see whether
women kept their various appointments. Compliance rates with subsequent biopsy
of lesions for women who received navigation were 100%.
The success of the navigator model is further illustrated by the Harlem Cancer
Education and Demonstration Project, which functioned as a patient support
program.14 The target group was low income, predominantly minority (over 90%
Black and Hispanic) individuals of low educational status. The Harlem Cancer
Education and Demonstration Project used a PN program to help address barriers
that underserved individuals have in trying to obtain follow-up services once an
abnormal breast cancer screening finding or an established cancer diagnosis was
made. The navigators in this model, although not volunteers, were individuals with
low salaries and limited training and experience who learned through on-the-job
training, monitoring, and program participation. The navigator support was initi-
ated at the time that the patient received abnormal test results, to ensure adequate
clinical follow up of these results. The navigators did not function as peer educators
and did not receive formal education. In this program, the navigator met with the
patient after referral, explored barriers faced by the patient, and reported problems
at staff meetings where possible solutions would be generated. The need for patient
navigation was so evident by the overwhelming demand and numerous referrals
that a second, bilingual PN was hired. Although there were no significant differences
between the two groups in whether they had a breast biopsy, 85.7% of those navi-
gated completed recommended biopsies, compared to only 56.5% of nonnavigated
patients, with the trend suggesting that navigation may have facilitated the completion
of the recommended biopsies.14 In addition, for those who completed biopsies, the
navigated patients completed the biopsy in significantly less time than the nonnavi-
Given these successes of patient navigation, and the paucity of data regarding
this approach for CRC screening, the main goal of this study was to test whether a
PN program, in addition to physician recommendation, could help overcome the
barriers that medically underserved people might have in trying to obtain clinical
CRC screening services. We hypothesized that a PN who is culturally sensitive to
the needs of the community would enhance participation in CRC screening over
and above that which can be accomplished by physician recommendation alone.
This study was designed as a prospective clinical trial intended to determine
whether a PN would enhance CRC screening participation beyond physician
recommendation alone in a neighborhood health care setting. Institutional Review
Board approval was obtained before conducting this study.
Participants were identified from among patients attending a primary care practice
in East Harlem, New York City, between January and May 2002. The practice is a
federally qualified heath center providing over 40,000 visits annually to over 9,000
PATIENT NAVIGATOR USE IN COLORECTAL CANCER SCREENING 219
individuals. It has a full range of primary care services and is staffed by permanent
physicians, ensuring continuity of care. Eligible subjects included men and women
aged 50 or older who (1) had not had an FOBT within the past year, (2) had not
had an FS or barium enema within the past 3–5 years, and (3) had not had a
colonoscopy within the past 10 years. Charts of all scheduled patients were successively
reviewed for eligibility, and the research assistant (RA) approached prospective
participants. One hundred twenty-five people were approached, and 88 (70%)
agreed to participate (Fig. 1). The 37 patients who refused to participate did not differ
by gender distribution from those who agreed. Of the 88 who agreed to participate,
4 were later found to be ineligible because of previous CRC screening and 6 were
lost to follow up (1 died and 5 medical charts could not be located). Thus, the study
group consisted of 78 presumably asymptomatic participants who were then ran-
domized into two groups: 38 people who received patient navigation services (PN+
group) and 40 who did not receive patient navigation services (PN− group). Inter-
views and informed consent were administered in either English or Spanish,
depending on the participant’s preference. At the time of recruitment, a 17-item
sociodemographic questionnaire was also completed.
FOBT cards were placed in the charts of all participants both to serve as a cue
to the physician and to ensure that the physician was blinded to the patient’s study
state. All participants were asked by their physician to complete FOBT cards and
were recommended to undergo endoscopic screening (either FS or colonoscopy).
We did not, however, assess whether all participants received a recommendation for
endoscopic screening from their physician. Participants in the PN+ group were then
Refused to participate
Agreed to participate
• Had previous CRC screening (n=4)
• Lost to follow-up (n=6)
Not navigated (“PN-”)
Selection and randomization of subjects.
220JANDORF ET AL.
assigned to the PN for assistance with completing the screening process, including
the FOBT cards and endoscopic procedures that had been recommended by their
physician (Fig. 2). The PN provided written reminders, telephone calls, and/or
scheduling assistance to the participants, encouraging participation in CRC screening.
One of the general functions of the PN was to provide support and advocacy for the
participant and assist with patient education regarding CRC risks, prevention, and
other relevant information. The PN coordinated with health care facilities to schedule
appointments for FS as per the physician’s request. The PN was from a similar
cultural background to most participants because we expected that similarity of
cultural patterns, values, experiences, and problems would likely enhance the
participant’s identification and comfort with the navigator.
All participants in the study group were followed to determine whether they indeed
had completed CRC screening tests. Completion of CRC screening tests was deter-
mined by a series of sequential chart reviews conducted in a nonblinded fashion by
the PN/RA (Fig. 2). Following the completion of the informed consent, the first
chart review confirmed the participant’s eligibility. At this time, FOBT cards were
placed in the reviewed charts. A second chart review took place 2–3 weeks after the
initial interview and medical visit. FOBT completion as well as referral for FS or
colonoscopy was recorded. Participants in the PN+ group who had not returned
their FOBT cards were also called during this period and reminded of their need to
complete their screening for CRC and given any assistance they needed. The third
chart review was performed 3 months after the initial contact. During this time,
FOBT completion as well as FS or colonoscopy completion was noted. After 6
Chart Review #1
• establish eligibility
• signed Informed Consent
• place FOBT cards in charts
• physician recommends FOBT & FS/colonoscopy
Chart Review #2
• was FOBT completed?
• was FS/colonoscopy referral made?
Chart Review #3
• was FOBT completed?
• was FS/colonoscopy completed?
Chart Review #4
• was FOBT completed?
• was FS/colonoscopy completed?
Schema of chart reviews.
PATIENT NAVIGATOR USE IN COLORECTAL CANCER SCREENING 221
months, the charts of all participants were reviewed for a fourth and final time to
assess FOBT, FS, and/or colonoscopy completion.
The RA served as the PN. She contacted the PN+ subjects 2–3 weeks after the initial
contact to educate the participant about any additional information on both screening
techniques and to begin navigation services. During the telephone call, the navigator
inquired whether the participant had a scheduled appointment for his or her endo-
scopic examination as well as the status of the FOBT card completion. This served
to identify those participants who made appointments before receiving navigator
services. If necessary, the PN reviewed the collection technique and return proce-
dure for FOBT cards. At this time, she addressed any barriers that the participant
described with regard to collecting the specimens. The PN and participant would
then set a date for when the participant intended to deliver the cards to his or her
physician. Two days after the scheduled date, the PN would contact the participant
again to determine whether the specimens had been returned to the physician and
checked. If necessary, the PN addressed any problems or concerns, including lost
cards or lack of understanding about the test.
If the participant did not schedule an appointment for FS, the PN reminded the
participant that the physician referred him or her for an FS. The PN would then ask
the participant whether he or she was interested in having an FS. If the patient said
“yes,” she or he was assisted in scheduling an appointment. If the participant said
“no,” the barriers that prevented that participant from completing the recom-
mended screening procedure were recorded. Once the appointment was scheduled,
the PN would call the participant 2 days before the FS to remind him or her of the
scheduled procedure. The PN addressed further concerns that the participant might
have regarding bowel preparation before the procedure, the protocol during the
procedure, and expectations after the procedure. A gastroenterologist (S.H.I.)
addressed any medically related concerns, problems, or questions. The PN called
the participant after the date of the scheduled screening procedure and inquired
whether he or she had the procedure done. If the procedure had not been
completed, the PN would assess the situation, and any barriers or problems were
discussed and resolved, if possible.
Data were entered and statistical analyses performed by using SPSS (version 11.0
for Windows). Descriptive statistics were generated on all study variables. Bivariate
analyses were conducted using the chi-square and Fisher exact tests for categorical
data. The Student unpaired t test was applied to continuous data. Interval level scale
variables were tested for homogeneity of variance. Findings with a P value <.05
(two-sided) were deemed statistically significant.
As summarized in Table 1, the PN+ and PN− groups did not differ by any sociode-
mographic characteristic. The mean age was 61.2 years. Similar to the demographics
of this clinic population, 74.4% of all participants were female, and 82.1% were
Hispanic (46.2% completed the assessments in Spanish). Approximately 29.5%
reported being married or living as married. Consistent with this low-socioeconomic
clinic population, only 11.5% had at least a high-school degree, two thirds had an
222 JANDORF ET AL.
annual income under $10,000, and fewer than 10% were employed, despite living
in the United States for an average of 37.3 years. Nonetheless, the vast majority had
a primary health care provider. Over two thirds had some form of public insurance
(Medicare, Medicaid, or both). Over one third reported a family history of cancer.
The findings for CRC screening adherence are summarized in Table 2. Following
their initial visit, at the 3-week chart review (during which no navigation services
had been provided), no significant differences were noted in FOBT completion
(26.3% vs. 17.5% in PN+ and PN− groups, respectively). However, at the 3-month
chart review, FOBT was completed by 42.1% of those in the PN+ group compared
with only 25% in the PN− group (P=.086). At that time, 18.4% of the PN+ had
made an endoscopy appointment, compared with none in the PN− group (P=.005),
and 15.8% of the PN+ group completed their endoscopic examination, compared
with only 5% of the PN− (approaches trend; P=.115). Importantly, by the time of
the final chart review 6 months after enrollment, 23.7% of the PN+ group had com-
pleted an endoscopic examination, compared with only 5% of the PN− group
Our rationale for pursuing this line of investigation is based on two observations.
First, traditional interventions that have relied purely on standardized educational
interventions have not substantially increased participation in cancer screening. Second,
TABLE 1. Sociodemographics of the study population
Variable PN+ group PN− groupOverallSignificance
Gender (% female)
Ethnicity (% Hispanic)
Language (% English)
Marital status (% married)
Education level (% at least high-school degree)
Annual income (% <$10,000)
Employed (% yes)
Years living in USA (mean±SD)
Have a primary care provider (% yes)
Have public health insurance (% yes)
Family history cancer (% yes)
40.6±17.6 34.2±20.4 37.3±19.2
TABLE 2.Adherence to screening
FOBT, fecal occult blood test.
Variable PN+ groupPN− groupSignificance
Completed FOBT after 3 weeks (before navigation) (% yes)
Completed FOBT after 3 months (% yes)
Had endoscopy appointment at 3 months (%)
Completed endoscopy at 3 months (%)
Completed endoscopy at 6 months (%)
PATIENT NAVIGATOR USE IN COLORECTAL CANCER SCREENING 223
PN programs have successfully enhanced screening for other types of cancers, as
described above. Because little is known about the utility of a PN system for CRC
screening, we decided to investigate this approach in patients who were recommended
by their physician to undergo CRC screening. Unlike screening guidelines for other
cancers, CRC screening guidelines permit the physician and patient to choose from
several alternatives. The American Cancer Society recommends that asymptomatic,
average-risk individuals over age 50 should undergo one of the following: (1) FOBT
annually, (2) FS every 5 years, (3) FOBT annually plus FS every 5 years, (4) colonos-
copy every 10 years, or (5) barium enema every 5 years. Although the availability to
choose among several screening options may appeal to some, it can also be a source
of confusion or lack of compliance on the part of the physician and patient.
We reasoned that if a PN system was effective, it might serve as a useful adjunct
to reduce the complexities of CRC screening. We focused our efforts on average-risk
subjects who were eligible for CRC screening. In keeping with the navigator studies
cited above, our patients were also predominantly minority (Hispanic), urban, and
of low socioeconomic status. By using a primary care clinic as our portal of entry,
we eliminated the barrier of subjects not having a physician.
Our findings demonstrate that within 6 months after physician recommendation
for CRC screening, a PN system was effective at ensuring that significantly more
patients received CRC screening by endoscopy. This finding is important because
endoscopic screening is considered superior to FOBT alone for reducing CRC mor-
tality. That significantly more patients in the PN+ group had an appointment for
endoscopy at 3 months further substantiates the usefulness of the PN approach. In
addition, the PN+ group showed a trend toward increased completion of FOBT
cards after 3 months, although this trend did not reach statistical significance. These
findings therefore suggest that in a predominantly poor, urban, minority population,
PNs can have a positive impact on CRC screening, with the potential for reducing
the mortality from CRC.
Despite this apparent success, it is worth noting several factors. First, although
the navigated group demonstrated higher rates of FOBT and endoscopic screening,
their rates were still quite low and not much higher than those reported in the
Behavioral Risk Factor Surveillance System study. Obviously, other barriers also
remain in this population. Other factors to consider in our study are the relatively
small sample size, the lack of documentation of physician recommendation, and the
use of one clinical site. Nonetheless, the higher rates of CRC test completion among
the PN+ group—despite the modest sample size—support the notion that a PN
approach will be a useful adjunct to enhance CRC screening among minority
In keeping with previous studies utilizing PNs, we selected a PN who was from
a similar ethnic background and who lived in the community from which we were
recruiting, so that she would be familiar with the community’s needs, organizations,
and leaders and would therefore be more culturally sensitive. Culturally sensitive
community-based approaches are more effective in minority populations than tradi-
tional approaches.15,16 Besides addressing issues of cultural sensitivity, our research
group is beginning to explore new approaches to overcoming organizational barriers
to endoscopic screening in this patient population.
This study was supported by the National Cancer Institute (U01-CA86107-10).
224 JANDORF ET AL. Download full-text
1. Greenlee RT, Hill-Harmonn MB, Murray T, Thun M. Cancer statistics, 2001. CA Cancer
J Clin. 2001;51:15–36.
2. Jemal A, Murray T, Samuels A, Ghafoor A, Ward E, Thun MJ. Cancer statistics, 2003.
CA Cancer J Clin. 2003;53:5–26.
3. Cancer Prevention and Control. Comparison of use of colorectal cancer tests with other
cancer screening tests, NHIS 2000. Available at: http://www.cdc.gov/cancer/colorctl/
calltoaction/slides_index.htm (slide35). Accessed April 11, 2005.
4. National Cancer Institute. SEER cancer statistics review 1973–1997. Available at: http://
seer.cancer.gov/csr/1973_1997/colorect.pdf. Accessed April 5, 2005.
5. Behavioral Risk Factor Surveillance System. Technical information and data. Available
at: http://www.cdc.gov/brfss/technical_infodata/surveydata/1997.htm. Accessed April 5,
6. Mandel JS, Bond JH, Church TR, et al. Reducing mortality from colorectal cancer by
screening for fecal occult blood. Minnesota Colon Cancer Control Study. N Engl J Med.
7. Mandel JS, Church TR, Bond JH, et al. The effect of fecal occult-blood screening on the
incidence of colorectal cancer. N Engl J Med. 2000;343:1603–1607.
8. Hardcastle JD, Chamberlain JO, Robinson MH, et al. Randomised controlled trial of faecal-
occult-blood screening for colorectal cancer. Lancet. 1996;348:1472–1477.
9. Burt RW. Colon cancer screening. Gastroenterology. 2000;119:837–853.
10. Newcomb PA, Norfleet RG, Storer E, Surawicz TS, Marcus PM. Screening sigmoidos-
copy and colorectal cancer mortality. J Natl Cancer Inst. 1992;84:1572–1575.
11. Selby JV, Friedman GD, Quesenberry CP Jr, Weiss NS. A case-control study of screening
sigmoidoscopy and mortality from colorectal cancer. N Engl J Med. 1992;326:653–702.
12. Tingen MS, Weinrich SP, Heydt DD, Boyd MD, Weinrich MC. Perceived benefits: a pre-
dictor of participation in prostate cancer screening. Cancer Nurs. 1998;21:349–357.
13. Frelix GD, Rosenblatt R, Solomon M, Vikram B. Breast cancer screening in underserved
women in the Bronx. J Natl Med Assoc. 1999;91:195–200.
14. Freeman HP, Muth BJ, Kerner JF. Expanding access to cancer screening and clinical
follow-up among the medically underserved. Cancer Pract. 1995;3:19–30.
15. Gauthier MA, Clarke WP. Gaining and sustaining minority participation in longitudinal
research projects. Alzheimer Dis Assoc Disord. 1999;13:S29–S33.
16. Stoy DB, Curtis RC, Dameworth KS, et al. The successful recruitment of elderly black
subjects in a clinical trial: the CRISP experience. Cholesterol Reduction in Seniors
Program. J Natl Med Assoc. 1995;87:280–287.