Correspondence

Screening for Colorectal Cancer

N Engl J Med 1995; 333:460-461August 17, 1995

Article

To the Editor:

Toribara and Sleisenger (March 30 issue)1 reviewed screening for colorectal cancer. Since 1992 prospective and case–control studies, such as that by Wahrendörf et al.,2 have shown a decreased mortality rate for colorectal cancer with the use of the Hemoccult II fecal occult-blood test. There is now strong evidence that it is possible to decrease mortality by screening with Hemoccult II persons older than 50 years of age. The evidence is as strong as that for mammography for breast-cancer screening, and stronger than that for Pap smears for cervical-cancer screening, for which there are no prospective studies. Since this knowledge could influence decisions on public health strategies, the French health minister established the French Working Group on Colorectal Cancer Screening last year.

The conclusion of Toribara and Sleisenger that “until better testing methods or more convincing data are available, fecal occult-blood tests for screening purposes should still be used at the discretion of individual physicians” is not supported by the data they discussed. The authors went on to state that “physicians . . . must weigh the benefit in terms of discovering cancer at an earlier stage and the consequent decrease in mortality against the immense costs of colonoscopies resulting from the low positive predictive value of the test.”

The authors provided only two points against a decision to screen for colorectal cancer. One was the fact that in the study by Mandel et al.,3 rehydration of the test slides led to a high positivity rate of 9.8 percent. In that study, without rehydration only 2.4 percent of the slides were positive and sensitivity only decreased from 92.2 percent to 80.8 percent. So, by dividing by 4 the number of colonoscopies, one can see that there is only a 10 percent decrease in sensitivity. This result dramatically reduces the “immense costs of colonoscopies.” Their second point was that the test has a “low positive predictive value.” This is astonishing because the positive predictive value of colonoscopies is no more than 1 to 2 percent in symptomatic patients, no more than 2 to 4 percent in patients with a family history, and 5 to 10 percent in patients with positive Hemoccult II tests — a result that in itself is one of the best indications for colonoscopy.

The conclusion of our working group is that a program of colorectal-cancer screening should be implemented in France, and this decision has been supported by the French health minister.

Gérard Dubois, M.D., M.P.H.
Centre Hospitalier Universitaire d'Amiens, 80054 Amiens, France

for the French Working Group on Colorectal Cancer Screening

3 References

1. 1

   Toribara NW, Sleisenger MH. Screening for colorectal cancer. N Engl J Med 1995;332:861-867
   Full Text | Web of Science | Medline

2. 2

   Wahrendorf J, Robra BP, Wiebelt H, Oberhausen R, Weiland M, Dhom G. Effectiveness of colorectal cancer screening: results from a population-based case-control evaluation in Saarland, Germany. Eur J Cancer Prev 1993;2:221-227
   CrossRef | Medline

3. 3

   Mandel JS, Bond JH, Church TR, et al. Reducing mortality from colorectal cancer by screening for fecal occult blood. N Engl J Med 1993;328:1365-1371
   Full Text | Web of Science | Medline

To the Editor:

Toribara and Sleisenger state that the ideal screening test for colorectal cancer would be “inexpensive, easy to administer, and highly specific and sensitive.” The test should also be acceptable to patients. Patients who undergo unpleasant medical procedures tend to be reluctant to follow subsequent recommendations. Even among highly selected patients — for example, those who undergo colonoscopy — as many as one in five fail to return for a subsequent examination.1 Indeed, patients' dissatisfaction is notorious for adversely affecting patients' behavior and limiting disease-prevention programs.2 Medical tests should be easy on patients as well as payers and clinicians.

Donald A. Redelmeier, M.D.
University of Toronto, Toronto, ON M4Y 1J3, Canada

2 References

1. 1

   Winawer SJ, Zauber AG, Ho MN, et al. Prevention of colorectal cancer by colonoscopic polypectomy. N Engl J Med 1993;329:1977-1981
   Full Text | Web of Science | Medline

2. 2

   Setaro JF, Black HR. Refractory hypertension. N Engl J Med 1992;327:543-547
   Full Text | Web of Science | Medline

Author/Editor Response

The authors reply:

To the Editor: Dr. Dubois states that screening for colorectal cancer in France (presumably using fecal occult-blood screening) has the support of both the French Working Group on Colorectal Cancer Screening and the French health minister.

In the interest of brevity, we did not cite the extensive literature debating the usefulness of fecal occult-blood tests in detecting polyps and early-stage cancers, since our intent was not to discuss these controversies. Rather, we wished to emphasize that recent studies involving both randomized or prospective and case–control designs have for the first time shown a benefit for colorectal-cancer screening in terms of mortality. Although the cost effectiveness of colorectal-cancer screening expressed as the cost per added year of life compares favorably with that of other screening methods,1 the benefit can be maximized only at considerable expenditure of medical resources (such as the cost involved in rehydrating Hemoccult slides). Thus, the controversy in colorectal-cancer screening has shifted from whether screening can reduce the mortality due to the disease to whether society is willing to bear the costs of such programs.

The decision of the French working group and the French health minister to recommend the implementation of large-scale screening for colorectal cancer is important both as an attempt to reduce mortality from a potentially fatal disease and in providing further necessary scientific data. However, we would caution that the reported variability in the sensitivity, specificity, and positive predictive value of fecal occult-blood testing makes projections of cost effectiveness uncertain.

Certainly, we agree with Dr. Redelmeier that patients' experiences during colonoscopy can markedly affect their willingness to follow subsequent recommendations. As screening for colorectal cancer gains wider acceptance, colonoscopy will necessarily become a more widely used diagnostic, therapeutic, and surveillance tool (perhaps as a primary screening method in certain circumstances). Data examining the reasons for patients' dissatisfaction with colonoscopy (i.e., discomfort during or after the procedure, patients' expectations, or confusion about the indications for the procedure or the procedure itself) are lacking at present and will play an important part in improving patients' acceptance of colonoscopy and thus the efficacy of any screening program.

Neil W. Toribara, M.D., Ph.D.
Marvin H. Sleisenger, M.D.
University of California, San Francisco, San Francisco, CA 94121

1 References

1. 1

   Office of Technology Assessment. Cost-effectiveness of colorectal cancer screening in average-risk adults. Background paper OTA BP-H-146. Washington, D.C.: Government Printing Office, 1995.
   

Citing Articles (3)

Citing Articles

1. 1

   Alain Braillon. (2010) Colorectal Cancer Screening: From Perspectives to Reality. Gastroenterology 139:3, 1065
   CrossRef

2. 2

   Jeff Gow. (1999) Costs of screening for colorectal cancer: An Australian programme. Health Economics 8:6, 531-540
   CrossRef

3. 3

   Ricardo N. Goes, Robert W. Beart, Anthony J. Simons, Leonard L. Gunderson, Gordon Grado, Oscar Streeter. (1997) Use of brachytherapy in management of locally recurrent rectal cancer. Diseases of the Colon & Rectum 40:10, 1177-1179
   CrossRef