Colorectal cancer is the fourth most common cancer in the United States and the second leading cause of cancer death. A person at age 50 has about a 5% lifetime risk of being diagnosed with colorectal cancer and a 2.5% chance of dying from it; ₃ the average patient dying of colorectal cancer loses 13 years of life.₄

More than 80% of colorectal cancers arise from adenomatous polyps. Although fewer than 1% of adenomatous polyps less than 1 cm will eventually develop into cancer, 10% of adenomatous polyps greater than 1 cm become malignant within 10 years, and about 25% become malignant after 20 years.₅ The prevalence of adenomatous polyps increases from 20% to 25% at age 50 to 50% by age 75-80.₆

Most colorectal cancers occur in persons at average risk, but 20% occur among patients with specific risk factors, such as those with a family history of colorectal cancer in a first-degree relative. A small proportion (6%) is associated with uncommon genetic syndromes such as familial adenomatous polyposis [FAP] or hereditary nonpolyposis colorectal cancer [HNPCC]. Other persons at increased risk include patients with longstanding ulcerative colitis, persons with previously diagnosed large adenomatous polyps or colorectal cancer, and those with a family history of adenomatous polyps diagnosed before age 60.

The USPSTF reviewed evidence of the effectiveness of the following screening tests for colorectal cancer: DRE, FOBT, sigmoidoscopy, colonoscopy, DCBE, and CT colography, singly and in various combinations.

Digital Rectal Examination/Office FOBT There is little evidence to determine the effectiveness of either DRE or a single office FOBT using a stool sample obtained on DRE. Fewer than 10% of colorectal cancers arise within reach of the examining finger, and some of these lesions will already be symptomatic. The sensitivity of a single office FOBT is likely to be substantially lower than that of screening protocols involving multiple test cards: in one study the first test card would have missed 42% of cancers detected by screening.₇ Samples collected by DRE may be affected by other limitations, including inadequate amount of stool or trauma from the exam.

Fecal Occult Blood Testing Sensitivity of FOBT screening varies with the testing protocol. Sensitivity and specificity of a single test have been estimated at 40% and 96% to 98%, respectively. Hydration of specimen increases sensitivity (60%) but reduces specificity (90%).₈ Of patients who have a positive FOBT using rehydrated slides, only 2% will have cancer; 6% to 8% will have cancer or a large polyp. Using unrehydrated specimens, 5% to18% of patients with a positive test will have cancer; 20% to 40% will have large polyps or cancer. The probability of cancer increases as the number of positive test windows increase. Tests that incorporate quantitative measures of heme and genetic stool markers have not been evaluated with respect to mortality reduction. Sensitivity and specificity change when screening is analyzed as a program of periodic screens. Annual screening with hydrated specimens detected 49% of all incident cancers, but 38% of all subjects had at least one colonoscopy due to positive results.₉ Programs using unrehydrated specimens and/or biennial testing detect a smaller proportion of cancers (27% to 39%) but require fewer colonoscopies (5% to 28%).₁₀-₁₁

Sigmoidoscopy First-time sigmoidoscopic screening detects approximately 7 cancers and about 60 large or high-risk polyps per 1,000 examinations.₁₂ Although sigmoidoscopy can only visualize the lower half of the colon,₁₃ it has been estimated to identify 80% of all patients with significant findings in the colon, because findings on sigmoidoscopy will trigger examination of the entire colon. It is difficult to quantify the “false-positive” rate of endoscopic screening, but screening may lead to the removal of many polyps that are of low malignant potential or that would not have caused clinical disease.

FOBT and Sigmoidoscopy Combining FOBT and periodic sigmoidoscopy has been advocated to improve the sensitivity of screening. In three recent randomized trials, performing flexible sigmoidoscopy in addition to FOBT yielded approximately 7 additional cancers or large polyps per 1,000 patients compared to FOBT alone.2 Adding FOBT did not improve the yield over sigmoidoscopy alone at the initial screening in these studies, which used flexible sigmoidoscopy; but did in an earlier study that used rigid sigmoidoscopy. Whether additional rounds of FOBT screening will have added benefits over flexible sigmoidoscopy has not been assessed.

Double Contrast Barium Enema Most studies of DCBE have important limitations for determining accuracy in an asymptomatic screening population. Previous studies have reported high sensitivity (86% to 90%) of DCBE for colorectal cancer and polyps, and high specificity (95%). In the National Polyp Study, however, DCBE detected only 48% of polyps greater than 1 cm.₁₄ Sensitivity might be higher in a typical screening population where the proportion of large polyps is higher. Specificity of DCBE in this study was 85%.

Colonoscopy Colonoscopy recently has been advocated for screening, usually at 10- year intervals or as a once-in-a-lifetime examination at age 55-65. The accuracy of colonoscopy is difficult to evaluate because it is usually considered the criterion standard. Estimated sensitivity of a single exam is 90% for large polyps and 75% for small polyps (less than 1 cm).₁₅ As with sigmoidoscopy, specificity is difficult to define. Many patients will have polyps detected or removed on colonoscopy, but only a minority of those would have developed cancer.

Computed Tomography (CT) Colography CT colography, or “virtual colonoscopy,” is a noninvasive procedure for producing images of the colonic lumen. The examination, which can be performed in 10 to 15 minutes, currently requires a preparation similar to colonoscopy, followed by installation of air through a rectal tube. Although CT colography can be relatively sensitive and specific in research settings (85% to 90%), recent reports have suggested lower accuracy when performed by less experienced examiners. Small and flat polyps are less well visualized on CT colography than are cancers and large polyps. Studies have not yet examined clinical outcomes with CT colography screening.

Fecal Occult Blood Testing Three randomized controlled trials (RCTs), all using the Hemoccult™ test kit, show reductions in risk of death from colorectal cancer from 15% to 33% from periodic FOBT screening. Two European trials, which randomized patients prior to agreement to participate and used biennial screening and unrehydrated test cards, found 15% to 18% reductions in mortality.10,11 In a U.S. study, which randomized volunteers and used rehydrated test cards, colorectal cancer mortality after 18 years of follow-up was 33% lower among persons advised to undergo annual FOBT than among controls who received usual care (9.46 versus 14.09 deaths per 1,000 patients screened); biennial screening reduced mortality by 21%.9,₁₆ A fourth trial conducted in Sweden has not reported final mortality results, but no significant mortality reduction was reported after 2 rounds of rehydrated testing (RR, 0.88; 95% CI, 0.69 - 1.12).

Sigmoidoscopy Current evidence of the effectiveness of sigmoidoscopy is limited to several well-designed case-control studies, but 2 ongoing RCTs of screening with flexible sigmoidoscopy are expected to report results within 5 years. A case-control study in a large health plan that had implemented rigid sigmoidoscopy screening suggested that screening reduced the risk of death from cancers within reach of the rigid sigmoidoscope by 59%.₁₇ A second case-control study in which 75% of the examinations were performed with a flexible instrument found similar protection.₁₈

FOBT and Sigmoidoscopy No RCTs have examined whether combining FOBT and sigmoidoscopy would lower mortality or morbidity more than either test alone. In a nonrandomized, controlled study involving more than 12,000 first-time attendees at a preventive-health clinic screened using rigid sigmoidoscopy, the addition of FOBT detected more cancers on initial screening than sigmoidoscopy alone, but mortality after 9 years was not significantly lower (0.36 per 1,000 patient-years in patients receiving both tests versus 0.63 per 1,000 patient years in controls; p = 0.11).₁₉ Whether results are generalizable to flexible sigmoidoscopy is uncertain.

Double Contrast Barium Enema No trial has examined the ability of screening barium enema to reduce the incidence or mortality from colorectal cancer.

Colonoscopy The effectiveness of colonoscopy to prevent colorectal cancer or mortality has not been tested in a randomized clinical trial. The National Polyp Study, a randomized trial of different intervals of surveillance after polypectomy, estimated that 76% to 90% of cancers could be prevented by regular colonoscopic surveillance exams.₂₀ These results should be interpreted with caution, however, because they are based on historical controls, and trial participants had more complete polyp removal than may occur in the screening setting. A single case-control study suggests that colonoscopy is associated with lower incidence of colon cancer (odds ratio = 0.47; 95% CI, 0.37-0.58) and lower mortality from colorectal cancer (odds ratio = 0.43;95% CI, 0.30-0.63).₂₁ Slightly greater benefits of colonoscopy have been predicted in models that project benefits based on sensitivity of screening and rates of polyp progression.

CT colography No studies have evaluated the effectiveness of CT colography in reducing morbidity or mortality from colorectal cancer.

There are few data to determine optimal age for starting or stopping screening. FOBT has been proven effective for persons aged 50-80 and sigmoidoscopy is associated with reduced mortality in persons older than 45. One cost-effectiveness model suggests that beginning screening at age 40 rather than at age 50 would offer less than a 1-day average improvement in life expectancy. Randomized trials suggest that a life expectancy of at least 5 years may be required to realize the benefits of screening.

FOBT has few potential harms but false-positive tests can lead to invasive procedures such as colonoscopy. Sigmoidoscopy can, in rare instances, lead to bowel perforation (1 to 2 per 10,000 examinations).₂₂ In a study of 1,235 screening sigmoidoscopies, adverse effects included pain (14%), anxiety, bleeding (3%), gas or flatus (25%), but no perforations.12 One patient died from complications after surgery to remove a severely dysplastic adenoma. A survey of barium enema experience reported that important complications of any type occurred in 1 in 10,000 examinations; perforation occurred in 1 in 25,000 examinations; death in 1 in 55,000 examinations.₂₃

Screening colonoscopy poses higher risks than FOBT or sigmoidoscopy, both because it is a more invasive procedure and because generally it is used with conscious sedation, which may lead to complications. The risks of colonsocopy depend on whether it is used simply for screening and diagnosis, or whether it is also used for therapeutic procedures (eg, removal of polyps). In two studies of screening colonoscopies in more than 5,000 patients, 0.2% to 0.3% had major complications during or immediately after the procedures, the most common being bleeding requiring hospitalization or emergency care.₂₄,₂₅

Risks are higher in therapeutic procedures ( eg, when polypectomy is performed ) than in diagnostic or screening procedures. Rates of perforation for diagnostic procedures in 16 published studies ranged from 0.03% to 0.61%. There are few data on bleeding complications but one study reported no bleeding events in 250 patients.2

The complication rates for therapeutic procedures were higher in some studies: 0.07% to 0.72% for perforations and 0.2% to 2.67% for bleeding. Death was rare (between 1 in 16,000 to 1 in 27,000) and more likely in symptomatic patients with acute problems or those with comorbid conditions. The mortality rate as a result of screening is likely to be on the lower end of this range. Complication rates could increase, however, if widespread adoption of colonoscopy leads to more procedures by less skilled endoscopists. Data are lacking on complications of CT colography.

Some patients report that they find the FOBT unpleasant or difficult to perform, but 50% to 70% of patients will complete FOBT when advised to by a clinician. A reminder system can increase adherence rates by an average of 14%. Studies conducted in primary care settings have found rates of adherence for sigmoidoscopy to be 25% to 50% for the initial test, but there are no data on adherence to repeat examinations. When given information about screening options and offered the choice of FOBT alone, sigmoidoscopy alone, or both tests together, most patients in an academic internal medicine clinic preferred both tests or FOBT alone; only 8% to 13% preferred sigmoidoscopy alone.₂₆ However, patient adherence to combined testing is lower than it is for sigmoidoscopy or FOBT alone. Patients' acceptance of barium enema screening has not been evaluated.

Studies examining the relative discomfort of barium enema and colonoscopy have produced inconsistent results. In one study of patients in a population with considerable previous screening experience, 38% preferred colonoscopy to other methods. The acceptability and feasibility of CT colography have not been examined.

Among 6 high-quality cost-effectiveness analyses examining only direct costs, the average cost-effectiveness ratio values for screening adults older than 50 with each of the major strategies were under $30,000 per life-year saved (Year 2000 dollars).2 Studies varied as to which strategy was most cost-effective, however.

The Task Force grades its recommendations according to one of 5 classifications (A, B, C, D, I) reflecting the strength of evidence and magnitude of net benefit (benefits minus harms):

A. The USPSTF strongly recommends that clinicians routinely provide [the service] to eligible patients. The USPSTF found good evidence that [the service] improves important health outcomes and concludes that benefits substantially outweigh harms.

B. The USPSTF recommends that clinicians routinely provide [the service] to eligible patients. The USPSTF found at least fair evidence that [the service] improves important health outcomes and concludes that benefits outweigh harms.

C. The USPSTF makes no recommendation for or against routine provision of [the service]. The USPSTF found at least fair evidence that [the service] can improve health outcomes but concludes that the balance of benefits and harms is too close to justify a general recommendation.

D. The USPSTF recommends against routinely providing [the service] to asymptomatic patients. The USPSTF found at least fair evidence that [the service] is ineffective or that harms outweigh benefits.

I. The USPSTF concludes that the evidence is insufficient to recommend for or against routinely providing [the service]. Evidence that the [service] is effective is lacking, of poor quality, or conflicting and the balance of benefits and harms cannot be determined.

Good: Evidence includes consistent results from well-designed, well-conducted studies in representative populations that directly assess effects on health outcomes.

Fair: Evidence is sufficient to determine effects on health outcomes, but the strength of the evidence is limited by the number, quality, or consistency of the individual studies, generalizability to routine practice, or indirect nature of the evidence on health outcomes.

Poor: Evidence is insufficient to assess the effects on health outcomes because of limited number or power of studies, important flaws in their design or conduct, gaps in the chain of evidence, or lack of information on important health outcomes.