Effective Clinical Practice
Joe V. Selby
For author affiliations, current addresses, and contributions, see end of text.
The quantity and quality of evidence supporting the effectiveness of colorectal cancer screening now rivals that of Pap smears and mammography screening. Table 1 synthesizes what we now know. Three large randomized trials were highly consistent in showing mortality reductions with screening fecal occult blood test (FOBT) and colonoscopic follow-up of positive results. (1-3) One small trial, (4) several case-control studies, (5-8) and one cohort study (9) were consistent in showing both incidence and mortality reductions with sigmoidoscopy screening.
Evidence for the effectiveness of the two other approaches, colonoscopy and double-contrast barium enema (DCBE), is less direct and rests primarily on the studies of FOBT and sigmoidoscopy. Given the relative length of the instruments, it is not surprising that colonoscopy is more sensitive than sigmoidoscopy. (10, 11) It seems highly plausible to extrapolate the proven benefits of sigmoidoscopy to the entire colon for colonoscopy, particularly when results of the three FOBT trials are considered (studies designed on the assumption that colonoscopic screening could lower colorectal cancer mortality rates). The goal of the FOBT is simply to make colonoscopic screening more efficient by identifying those most likely to benefit. The evidence for DCBE is limited to descriptive studies showing that it has a relatively high sensitivity (50% to 94%) when compared with endoscopy for cancer and larger adenomatous polyps. (12, 13) Compared with FOBT, DCBE's sensitivity is much higher for adenomas and at least as high for cancer. It therefore seems plausible that screening with DCBE should also provide benefit when positive results are followed up with endoscopic polypectomy or surgery.
In this issue of ecp, Erban and colleagues (14) show that, despite this evidence, levels of colorectal cancer screening remain well below those for Pap smear and mammography. In this 1998 survey of Massachusetts residents, only 51% of adults aged 50 years and older reported being current with any form of colorectal screening. The authors' approach, of reporting on combined exposure to any of the several choices for screening, is highly useful. It is consistent with current guidelines, as cited by the authors, which recommend that patients and physicians select from a menu of effective screening options; it also provides the best assessment of the coverage of a population by an effective screening method.
The authors point out that having a variety of approaches, which differ in recommended screening intervals, complicates measurement of screening performance by providers. This may be unfortunate, in that systematic measurement and reporting has proven to be an effective means of raising mediocre performance levels for other prevention measures. However, deficits in available evidence and even greater deficits in the widespread availability of screening continue to preclude designation of a single, "best," national approach to colorectal cancer screening.
With respect to the evidence, uncertainty about the relative benefits and risks of the available approaches is an important reason to preserve the menu of screening options. Flexible sigmoidoscopy and FOBT screening appear to give roughly equivalent mortality reductions. For sigmoidoscopy, 60% to 90% of screened patients benefit, but only for the 50% to 60% of tumors that lie within reach of the sigmoidoscope. For FOBT, clinical trials find a 15% to 33% mortality reduction for patients with cancer anywhere in the colon and rectum. These reductions are achieved despite substantial noncompliance with screening and colonoscopic follow-up of positive results in the trials. It is reasonable to expect that persons who comply with both screening and follow-up would enjoy a somewhat greater reduction in risk. We do not know whether or by how much combining FOBT with flexible sigmoidoscopy could further reduce mortality. Some added benefit seems probable in that sigmoidoscopy should discover many adenomas and early cancers that do not bleed, and FOBT should confer some added protection against proximal tumors unaccompanied by distal marker lesions discoverable at sigmoidoscopy. However, this added benefit may be quite small. Empirical studies are needed to establish and quantify any benefit and to determine if the combination is cost-effective and acceptable to patients.
Arguments for colonoscopy as a primary screening strategy for adults at average risk have been advanced by two recent studies (10, 11) that showed, not surprisingly, that screening colonoscopy detects between 27% and 47% more important pathologic conditions than a flexible sigmoidoscopy screening strategy. Again, the absolute gain may be small, particularly for younger and lower-risk persons, and the safety and acceptability of primary colonoscopy screening are yet to be fully established.
A second, no less critical, reason to maintain the current variety of screening options is the state of the screening infrastructure in the United States. Persistently low levels of screening with sigmoidoscopy are undoubtedly due, in part, to the complexities of arranging screening, either in the primary care physician's office or through referral to a specialist. Low reimbursement rates for screening sigmoidoscopy (15) are a further deterrent to active endoscopic screening. Large-scale referral for screening colonoscopy would very quickly overwhelm available resources in nearly all U.S. settings, regardless of reimbursement rates. A deficit of skilled and willing colonoscopists would quickly lead to long queues, and it is unclear whether enough people could be recruited and trained to perform these examinations. Large endoscopic screening units, staffed by nonphysician endoscopists and supervised by gastroenterologists, may ultimately provide a safe and affordable solution. However, very few of these exist today, (16) and there are none for colonoscopy.
On the brighter side, a 51% current screening level is far from no screening and these figures appear similar to national rates. (17) Colorectal cancer incidence has been declining in the United States since 1985, (18, 19) with earlier and larger decreases for distal than for proximal cancer. Some have suggested that this pattern is due, at least in part, to the effects of screening. (20-22) The data presented by Erban and coworkers lend some credence to these suggestions. Given the relatively high effectiveness of colorectal screening, exposure of half the population could reasonably be expected to lead to detectable decreases in cancer incidence. The surprisingly high figure of a 20% lifetime exposure to colonoscopy, often done after a positive FOBT or sigmoidoscopy, may well contribute to the modest decreases in proximal cancer.
For the present, we are fortunate to have several effective strategies for colorectal screening because it gives alternatives to patients, who differ in their subjective and objective risk for colorectal cancer and in their willingness to tolerate screening tests, and to primary care physicians, for whom availability of the screening options varies widely. Until the superiority and feasibility of a single approach to screening is clearly demonstrated, there is good reason to guard against efforts to prematurely anoint a single strategy. Efforts to further increase colorectal cancer screening are needed. For now, these efforts should assess the screening resources available in specific settings and encourage patients to choose from the available alternatives. Meanwhile, the search for an optimal approach to colorectal cancer screening, in terms of effectiveness, safety, acceptability, and costs, should continue.
1. Mandel JS, Bond JH, Church TR, et al. Reducing mortality from colorectal cancer by screening for fecal occult blood. Minnesota Colon Cancer Control Study. New Engl J Med. 1993;328:1365-71.
2. Hardcastle JD, Chamberlain JO, Robinson MH, et al. Randomised controlled trial of faecal-occult-blood screening for colorectal cancer. Lancet. 196;348:1472-7.
3. Kronborg O, Fenger C, Olsen J, Jorgensen OD, Sondergaard O. Randomised study of screening for colorecatal cancer with faecal-ocult-blood test. Lancet. 1996;348:1467-71.
4. Thiis-Evensen E, Hoff GS, Sauar J, Langmark F, Majak BM, Vatn MH. Population-based surveillance by colonoscopy: effect on the incidence of colorectal cancer. Telemark Polyp Study I. Scand J Gastroenterol. 1999 Apr;34:414-20.
5. 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;326:653-7.
6. Newcomb PA, Norfleet RG, Surawicz TS, Storer BE, Marcus PM. Screening sigmoidoscopy and colorectal cancer mortality. J Natl Cancer Inst. 1992;84:1572-5.
7. Muller AD, Sonnenberg A. Protection by endoscopy against death from colorectal cancer. A case-control study among veterans. Arch Intern Med. 1995;155:1741-8.
8. Muller AD, Sonnenberg A. Prevention of colorectal cancer by flexible endoscopy and polypectomy. A case-control study of 32,702 veterans. Ann Intern Med. 1995;123:904-10.
9. Kavanagh AM, Giovannucci EL, Fuchs CS, Colditz GA. Screening endoscopy and risk of colorectal cancer in United States men. Cancer Causes Control. 1998; 9:455-62.
10. Lieberman DA, Weiss DG, Bond JH, Ahnen DJ, Garewal H, Chejfec G. Use of colonoscopy to screen asymptomatic adults for colorectal cancer. Veterans Affairs Cooperative Study Group 380. N Engl J Med. 2000;343:162-8.
11. Imperiale TF, Wagner DR, Lin CY, Larkin GN, Rogge JD, Ransohoff DF. Risk of advanced proximal neoplasms in asymptomatic adults according to the distal colorectal findings. National Polyp Study Work Group. N Engl J Med. 2000;343:169-74.
12. Winawer SJ, Steward ET, Zauber AG, et al. A comparison of colonoscopy and double-contrast barium enema for surveillance after polypectomy. N Engl J Med. 2000;342:1766-72.
13. Glick SN, Ralls PW, Balfe DM, et al. Screening for colorectal cancer. American College of Radiology. ACR Appropriateness Criteria. Radiology. 2000;215 Suppl:231-7.
14. Erban S, Zapka J, Puleo E, Vickers-Lahti M. Colorectal cancer screening in Massachusetts: measuring compliance with current guidelines. Eff Clin Pract. 2001;4:10-17.
15. Lewis JD, Asch DA. Barriers to office-based screening sigmoidoscopy: does reimbursement cover costs? Ann Intern Med. 1999;130:525-30.
16. Palitz AM, Selby JV, Grossman S, et al. The Colon Cancer Prevention Program (CoCaP): rationale, implementation, and preliminary results. HMO Pract. 1997;11:5-12.
17. Screening for Colorectal Cancer--United States, 1997. MMWR Morb Moral Wkly Rep. 1999;48:116-21.
18. Ries LAG, Wingo PA, Miller DS, et al. The annual report to the nation on the status of cancer, 1973-1997, with a special section on colorectal cancer. Cancer. 2000;88:2398-424.
19. Chu KC, Tarone RE, Chow W-H, Alexander GA. Colorectal cancer trends by race and anatomic subsites, 1975 to 1991. Arch Fam Med. 1995;4:849-56.
20. Cress RD, Morris CR, Wolfe BM. Cancer of the colon and rectum in California: trends in incidence by race/ethnicity, stage, and subsite. Prev Med. 2000;31:447-53.
21. Inciardi JF, Lee JG, Stijnen T. Incidence trends for colorectal cancer in California: Implications for current screening practices. Am J Med. 2000;109:277-81.
22. Selby JV. Explaining recent declines in colorectal cancer incidence: was it the sigmoidoscope? Am J Med. 2000;109:332-4.
23. Winawer SJ, Stewart ET, Zauber AG, et al. A comparison of colonoscopy and double-contrast barium enema for surveillance after polypectomy. N Engl J Med. 2000;342:1766-72.
Joe V. Selby, MD, MPH, Division of Research, Kaiser Permanente, 3505 Broadway, Oakland, CA 94611; telephone: 510-450-2106; fax: 510-450-2073; e-mail: email@example.com.