Learning Center
Plans & pricing Sign in
Sign Out

Title of GuidelineColorectal cancer screening in adults


									                                    Kaiser Permanente
                  Colorado Region : Colorectal Cancer Screening in Adults
 Date: July 25, 1997

 Responsible Party: Eric France MD MSPH, Ned Calonge MD MPH
 Approval: Departments of Preventive Medicine, Gastroenterology, Internal medicine, Family
 Medicine, Obstetrics/Gynocology.
 Review Dates: November 1997                              Next Review Date: August 2003
               November 1999
               August 2001

Title of Guideline: Screening: Colorectal Cancer Screening in Adults.

Rationale for Guideline: The U.S. Preventive Services Task Force supports colorectal cancer screening
in all persons aged 50 and over ("B" recommendation) using either fecal occult blood testing (FOBT) or
flexible sigmoidoscopy (SIG). Given the lack of evidence in support of one method of screening over the
other, the Regional Prevention Committee has recommends either every 10 year SIG or annual FOBT
screening for members of average risk for CRC.

Target Population: All members between the ages of 50 and 75 years. Also included in the target
population are members age 40 years and older who may be at increased risk for CRC because of a
family history of CRC or the existence of a predisposing medical condition.

Source of Evidence:

Guide to Clinical Preventive Services (1996), the U.S. Preventive Services Task Force and the current
medical literature (see reference list).

Settings for Application: Internal Medicine, OB/GYN, and Family Practice departments.

Method(s) for Measuring Compliance: The proportion age-eligible members screened with either SIG or
FOBT will be measured using the laboratory database and CIS procedure codes. Also, the proportion of
hemoccult cards returned and the FOBT positivity rate will be measured as will the follow-up rate of
positive test results as monitored by each department.

Responsible Party: Ned Calonge, M.D., M.P.H., Regional, Preventive Medicine, 344-7380
                        Guidelines for Routine Colorectal Cancer Screening

These guidelines are informational only and are not intended or designed to substitute the
reasonable exercise of independent clinical judgment by providers in any particular set of
circumstances for each patient encounter. The guidelines are flexible and are intended to be used
as a resource for integration with a sound exercise of clinical judgment. They can be used to
create an approach to care that is unique to the needs of each individual patient.


Screening for colorectal cancer (CRC) is recommended for all persons aged 50 to 75. For
members at average risk for CRC, annual screening with either flexible sigmoidoscopy (SIG)
screening every 10 years or non-rehydrated fecal occult blood testing annually is recommended.


Colorectal cancer (CRC) is the second most common form of cancer and has the second highest
mortality rate, accounting for over 150,000 new cases and about 57,000 deaths each year.
Recommendations (1996) by the US Preventive Services Task Force (USPSTF) support
screening for colorectal cancer in all persons aged 50 or over (“B” recommendation*) using either
fecal occult blood testing (FOBT) or sigmoidoscopy screening (SIG)1.

In 1995, our region’s colorectal cancer screening task force (CRCTF) reviewed the evidence in
support of CRC screening and concluded that there is evidence to support the effectiveness of
CRC screening. The Task Force at that time chose not to recommend periodic SIG screening,
given the start-up costs of such a program and the absence of randomized controlled trial data
demonstrating an increased effectiveness of SIG screening on CRC mortality compared to FOBT.
 This document will describe the evidence in favor of colorectal cancer screening and the
rationale behind the recommendations of the CRCTF.


Fecal Occult Blood Testing

Three large randomized controlled trials (RCTs) have been completed evaluating the effect of
FOBT screening on CRC mortality. The first of these trials, the Minnesota Colon Cancer
Control Study2 (MCCCS) found a 33 percent reduction in CRC mortality in the screened group
versus the non-screened group. Screening reduced mortality when the test was performed on an
annual basis: biennial (every two year) testing did not significantly reduce CRC mortality.

Because the MCCCS used rehydrated Hemoccult slides, the positivity rate for screening was high
(9.8 percent), generating a large volume of diagnostic evaluations: during the 13 year study
period, 38 percent of members in the FOBT cohort underwent a colonoscopy exam. This led to

 B Recommendation: There is fair evidence to support the recommendation that the condition be specifically
considered in a periodic health examination.

the speculation that the protective effect found with annual FOBT screening did not arise from its
ability to identify early bleeding CRCs, but rather from the high cumulative frequency of
colonoscopy which accompanies the chance finding of a positive test.

Two European CRC screening trials have since demonstrated the efficacy of screening with non-
rehydrated Hemoccult cards: The Lancaster, England(FT) and Fuhnen, Denmark(FT) studies
found reductions in CRC deaths of 15 and 18 percent, respectively, with biennial screening. In
both studies, a small proportion of subjects (<5%) underwent colonoscopy evaluation.

Together, these three RCTs demonstrate a reduction in CRC mortality with either annual or
biennial FOBT screening. No differences in all-cause mortality was seen in any of these studies:
although persons undergoing screening were less likely to die from CRC, their overall death rates
were no better than that of the unscreened population.


There have been no randomized controlled trails evaluating the screening efficacy of
sigmoidoscopy completed to date. The only controlled studies published are two case-control
studies4,5. The first of these studies4 showed that only 8.8 percent of the 261 case subjects
(persons who had died from CRC) had undergone sigmoidoscopy, versus 24.2 percent in the
control group (adjusted odds ratio = 0.41). The apparent benefit was just as strong when the
most recent sigmoidoscopic exam had occurred 9 to 10 years before, as when the examinations
were more frequent. A separate analysis in this study of a second cohort of 268 cases (matched
to controls) with CRC above the reach of the rigid sigmoidoscope demonstrated no benefit of the
screening test. The authors concluded that the use of the 60cm FSIG should lead to a reduction
of at least 30 percent in total mortality from colorectal cancer. A second case-control study has
since demonstrated similar results.

The long-term risk of CRC after excision of rectosigmoid adenomas has been evaluated in a
cohort study6 in which 1,618 patients were followed for an average of 13.8 years. The authors
found the risk of subsequent rectal cancer after the removal of rectal adenomas to be no higher
than that in the general population. The risk for colon cancer among patients with small (<1 cm)
tubular tumors (either single or multiple), was also no higher than that in the general population.
By contrast, the risk of colon cancer in patients with tubulovillous, villous, or large (1 cm)
tumors was 3.6 times greater than the incidence expected for this population. The results of this
study, coupled with the findings of a protective effect from flexible sigmoidoscopy lasting as
long as ten years in the case-control study, suggest surveillance colonoscopy following
polypectomy may be performed once every ten years.

Other Possible Methods of CRC Screening

No controlled studies have been performed to date to prospectively evaluate the use of either
60cm flexible sigmoidoscopy, colonoscopy, or air-contrast barium enema as a method of
screening for colorectal cancer in a controlled trial. Lieberman, et al recently found that
colonoscopy did detect about 20% more significant lesions (in the proximal colon) than would
have been expected to have been found with SIG alone. However, colonoscopy was associated
with a 3/1,000 serious complication rate, about 60 times the reported rate for SIGs. There is

insufficient data to conclude whether the risk/benefit comparison would favor a colonoscopy vs.
a SIG-based screening program. A number of modeling studies have been performed in order to
compare these screening modalities, and their results reflect the assumptions that go into the
model design. It was the decision of the CRCTF to focus only on methods for which direct
evidence presently exists to support CRC screening rather than on modeling efforts.

Screening Adherence

All cancer screening tests are limited by the degree to which their use is adopted by the general
public; even the most sensitive cancer screening test will fail if few people choose to participate.
 Estimates from the 1992 National Health Interview Survey (NHIS) found the prevalence of
screening during the previous three years for FOBT and SIG to be 26% and 9%, respectively9. In
order to increase the penetration of CRC screening, methods to improve adherence must be

A study evaluating the effect of a shared decision making approach on CRC screening adherence
was performed at the Westminster facility in 1996-1997. This study found that member
adherence to SIG screening was much lower than for FOBT screening (13% vs 24%). Providing
members with an information sheet comparing the two forms of screening did not improve
adherence: only 14% of persons in the shared decision making group completed screening.
Almost two-thirds of persons in this group chose to undergo FOBT screening rather than SIG
screening. The case-finding approach as seen in this study is unlikely to generate adherence
levels with screening above 25%, nor is it likely to generate great demand for SIG screening


The CRCTF recommends CRC screening continue to be performed on a case finding basis, in
which it is offered during routine health maintenance visits and, at the discretion of the caregiver,
during acute appointments, rather than expanding to use population-based methods of recruiting
member participation such as letters, telephone calls, or advertising. With three RCTs completed
evaluating FOB testing, there is strong evidence that CRC mortality is reduced with screening.

I. Recommendations for Average Risk Individuals

Screening for colorectal cancer (CRC) is recommended for all persons aged 50 to 75. For
members at average risk for CRC, annual screening with either flexible sigmoidoscopy (SIG)
screening every 10 years or non-rehydrated fecal occult blood testing annually is recommended.
Persons with a normal SIG exam (or colonoscopy) do not need repeat screening - either with
FOBT or SIG - for ten years. . Members who are offered CRC screening should be provided
with information about the test, including the expected benefits and harms. Members undergoing
screening should be counseled by their provider on the necessary restrictions in diet and
medication. If members indicate they choose not comply with the recommendation for SIG or
the return of the hemoccult cards, the provider should document that screening was offered and
the member chose not to participate. Asymptomatic patients greater than age 75 should not be
offered CRC screening on a routine basis; screening should not be performed if life expectancy
is less than 5-10 years or if the patient is too frail or debilitated to undergo colonoscopy or

surgery should a test be positive. Members who have a negative diagnostic work-up
(colonoscopy or barium enema/FSIG) do not need to have annual FOBTs for another 5-10 years.

II. Recommendations for Increased Risk Individuals

A.     Asymptomatic members, age 40-75, at increased risk (table 1) should be offered the
       opportunity to have a visual screening examination of the colon every 5 years.

                   Table 1. Increased risk conditions for colorectal cancer

                   Two or more first-degree relatives (parents or siblings) with CRC in a member
                   under the age of 60*.

                   A single first-degree relative with CRC diagnosed at or before age 55 in a
                   member under the age of 60*.

                   Previous large ( 1 cm) adenoma or adenoma with villous architecture.

                   Inflammatory Bowel Disease.

                   Hereditary nonpolyposis colorectal cancer (HNPCC).
                   Adenomatous polyposis syndromes.
                After the age of 60, family history of CRC no longer increases CRC risk.

B.     Members should be provided information about the procedures, including their expected
       benefits and harms, and invited to decide whether they would like to have a procedure.
       Based on their expected benefits, procedures should be recommended in the following
               1.     Because it has the greatest predicted benefit, colonoscopy should be
                      offered first.
               2.     If a patient declines to receive colonoscopy, barium enemas with a flexible
                      sigmoidoscopy should be offered as a second choice.
               3.     If a patient declines both a colonoscopy and a barium enema, flexible
                      sigmoidoscopy every 5 years should be strongly encouraged.

C.     Patients who decline all of the above tests may be offered an annual non-rehydrated
       FOBT. Members who are offered FOBTs should be provided with information about the
       test, including the expected benefits and harms. Members should have the benefit of a
       discussion with a provider prior to being given FOBT cards for use.

D.     Family history of CRC as described in table 1 raises the risk of CRC in a member under
       the age of 60. Members with such a family history who are age 60 or over have no
       increased risk compared to persons 60 or over without a family history of CRC8. A
       reasonable approach to CRC screening for members under age 60 with a family history of
       CRC may be to perform a colonoscopy and, if polyps are found, repeat the colonscopy
       every five years until the age of 60. If no polyps are identified at the time of the first
       colonoscopy, repeat visualization may not be necessary for another ten years.

Fecal Occult Blood Testing

A successful FOBT screening program requires the following:

1.     FOBTs are done on asymptomatic members between the ages of 50-75 years. Members
       presenting with bleeding per rectum should not be screened, but rather undergo the
       appropriate diagnostic evaluation.

2.     Compliance with bowel prep (no red meat, no raw fruits, etc) should be stressed by the
       provider when recommending FOBT screening, and again repeated by the lab technician
       when the stool cards are handed out. An education sheet will be given to the member by
       the lab technician along with the stool cards. This education effort will help reduce the
       number of false positive FOBTs. If a patient has a positive FOBT but did not comply
       with bowel prep, the screening test should nevertheless be considered positive.
       Repeating FOBTs among members whose bowel prep was inadequate is not
       recommended. Since CRCs bleed intermittently, it is possible that a subsequent negative
       FOBT after appropriate bowel prep reflects the intermittency of CRC bleeding.

3.     Providers should emphasize the importance of mailing back the stool cards within 2-4
       days after taking the first sample. A significant proportion of positive Hemoccult results
       will become negative if the completed slides are left at room temperature for 2-4 days9.

4.     Members on chronic aspirin or other non-steroidal anti-inflammatory drugs should refrain
       from using these medicines for seven days prior to and during the test period. Members
       taking a once daily dose of aspirin (either baby aspirin or 325 mg tablet) do not need to
       discontinue this regimen during FOBT screening, since a false positive result is unlikely
       to occur with low dose aspirin use. Members on coumadin should, if possible,
       discontinue this medication for seven days before undergoing FOBT screening. Those
       members who cannot be taken off of coumadin may choose to undergo barium enema
       screening every ten years, although there is no evidence in support of this form of CRC

5.     High-risk members who are in a colonoscopy surveillance program do not need FOBT
       screening. Average-risk members who have a negative diagnostic workup of a positive
       FOBT do not need FOBT screening during the subsequent 5-10 years.

Work-up of a Positive FOBT:

The first choice of a diagnostic work-up of a positive FOBT is colonoscopy. Evaluation of a
positive screening FOBT with colonoscopy should be completed within three months of the test
date. If the GI department is overwhelmed with requests for this procedure, or if the physician
and patient prefer to work-up the positive FOBT at a hub facility, the second choice would be
double contrast barium enema plus flexible sigmoidoscopy. Members who have a negative
diagnostic work-up do not need to have annual FOBTs for another 5-10 years.

Measurement of Compliance with Guideline

Compliance with CRC screening will be monitored by:

      1.      Measuring the proportion of people in the target age category screened with either
              SIG or FOBT

      2.      Measuring the proportion of hemoccult slides given out by the lab which are
              returned for analysis.

      3.      Measuring the proportion of FOBTs which are positive. This should reflect
              member compliance with dietary and medication restrictions.

      4.      The completeness of follow-up of members with a positive FOBT will be
              monitored by each facility.

      5.      Tumor registry data will be followed to look for shifts in the stage of disease at
              the time of diagnosis over time.


1.    Howe HL, Wingo PA, Thun MJ, Ries LAG, Rosenberg HM, Feigal EG, Edwards BK. Annual report to the
      nation on the status of cancer (1973 through 1998), featuring cancers with recent increasing trends. JNCI
      2001; 93:824-842.
2.    U.S. Preventive Services Task Force. Guide to clinical preventive services, 2 nd ed. Baltimore: Williams &
      Wilkins, 1996.
3.    Mandel JS et al. Reducing mortality from colorectal cancer by screening for fecal occult blood. Minnesota
      Colon Cancer Control Study. NEJM 1993; 328: 1365-1371.
4.    Lang CA, Ranshohoff DF. Fecal occult blood screening for colorectal cancer: is mortality reduced by
      chance selection for screening colonoscopy? JAMA 1994; 271: 1011-1013.
5.    Selby JV, Friedman GD, Quesenberry CP, Weiss NS. A case-control study of screening scimoidoscopy and
      mortality from colorectal cancer. NEJM 1992; 326: 653-657.
6.    Newcomb PA, Norfleet RG, Surawicz TS, Storer BE. Sigmoidoscopy and colorectal cancer mortality. Am
      J Epidemiol 1989; 130: 827. (abstract)
7.    Atkin WS, Morson BC, Guzick J. Long-term risk of colorectal cancer after excision of rectosigmoid
      adenomas. NEJM 1992; 326: 658-662.
8.    Centers for Disease Control, National Center for Chronic Disease Prevention and Health Promotion,
      Behavioral Risk Factor Surveillance System, 1999. http://apps.nccd,
9.    Simon JB. Occult blood screening for colorectal carcinoma: a critical review. Gastroenterology 1985; 88:
10.   Grossman S. A new era in colorectal cancer screening and surveillance. The Permanente Journal 1998.
11.   Colorectal Cancer Screening. Summary, Evidence Report: Number 1. Agency for Health Care Policy and
      Research, Rockville, MD.
12.   Basson MD, Panzini L, Palmer RH. Effect of nabumetone and aspirin on colonic mucosal bleeding time.
      Aliment Pharmacol Ther 200l; 15:539-542.


To top