Correspondence

Role of Fecal Occult-Blood Testing

N Engl J Med 1998; 339:774-775September 10, 1998

Article

To the Editor:

Simon's argument against population screening for colorectal cancer by fecal occult-blood testing (April 16 issue)1 contains inaccurate conclusions related to our Minnesota screening trial.2 First, he concludes that the fecal occult-blood test has limited sensitivity. An analysis of the sensitivity of individual screening tests and of repetitive screening in our trial shows that of the cases detected among subjects in compliance with the requirements of the test, only about 10 percent were missed at the time of the screening.3 Although only half the cancers diagnosed during the trial were detected by screening, only about 50 percent of the person-years in the study were covered by screening. The majority of interval cancers occurred in those not in compliance (the average compliance rate was about 75 percent) or during a four-year hiatus when no screening was conducted. A screening test does not necessarily have to be highly sensitive to be effective, as long as its repetitive use reliably detects a neoplasm before it becomes incurable.3

Second, Simon states that the specificity of fecal occult-blood testing results in unnecessary colonoscopies and that modification of the tests has not improved their performance. However, it is important to maximize the sensitivity of screening tests in order to detect most potentially fatal cancers, even if it causes false positive results. Unlike screening for some other cancers, a false positive fecal occult-blood test is not without value. It results in a colonoscopy — a test many now promote for primary screening; it provides reassurance about the risk of this common cancer; and it obviates the need for further screening for up to 10 years.

Third, Simon concludes that screening by fecal occult-blood testing is not cost effective. Although the cost of screening everyone would be high, it must be compared with the cost savings derived from detecting curable cancers and preventing cancers by resecting polyps. Using data from the Minnesota trial, economists calculated the cost of annual screening by fecal occult-blood testing to be $13,581 per year of life gained (not $35,000 to $40,000, as claimed by Dr. Simon).4

Finally, Simon criticizes screening by fecal occult-blood testing because current compliance is low. It is not surprising that the studies he cites showed poor compliance, since they preceded the publication of the randomized studies showing efficacy. Now that virtually all guidelines in the United States recommend screening, educational campaigns to improve compliance are under way. Of all the options for screening, compliance with fecal occult-blood testing may be the easiest to achieve. In the Minnesota trial, for example, compliance with both repetitive screening and diagnostic evaluations exceeded 75 percent over a period of 15 years.

John H. Bond, M.D.
Jack S. Mandel, Ph.D.
Timothy R. Church, Ph.D.
University of Minnesota, Minneapolis, Minneapolis, MN 55455

4 References

1
Simon JB. Should all people over the age of 50 have regular fecal occult-blood tests? Postpone population screening until problems are solved. N Engl J Med 1998;338:1151-1152
Full Text | Web of Science | Medline

2
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[Erratum, N Engl J Med 1993;329:672.]
Full Text | Web of Science | Medline

3
Church TR, Ederer F, Mandel JS. Fecal occult blood screening in the Minnesota study: sensitivity of the screening test. J Natl Cancer Inst 1997;89:1440-1448
CrossRef | Web of Science | Medline

4
Wagner JL, Tunis S, Brown M, Ching A, Almeida R. Cost-effectiveness of colorectal cancer screening in average-risk adults. In: Young GP, Rozen P, Levin B, eds. Prevention and early detection of colorectal cancer. London: W.B. Saunders, 1996:321-56.

Author/Editor Response

Dr. Simon replies:

To the Editor: Bond and his colleagues raise important and valid points, but our perspectives on fecal occult-blood screening still differ.

Sophisticated analyses of the Minnesota trial have reached disparate conclusions about sensitivity1,2; in part, this reflects the conceptual distinction between the sensitivity of the individual test and the overall sensitivity of the screening program. Although I agree that even a relatively insensitive test can be effective if used repetitively, the broad literature clearly indicates that fecal testing misses a substantial proportion of cancers and the vast majority of neoplastic polyps.3

Furthermore, enhanced sensitivity comes at the price of weak specificity. In the Minnesota program, a mere 2 percent of positive hydrated occult-blood tests were positive owing to cancer (i.e., 98 percent were false positives).4 This extreme inefficiency imposes on the public an inordinate number of needless invasive colonic workups and is a huge waste of both human and economic resources. It is true that a normal colonoscopy precipitated by a false positive test obviates further screening for up to a decade, but surely it is much better not to have a false positive result in the first place. Bond et al. argue that it is important to maximize sensitivity, but I think it is equally or even more important to maximize specificity. I hope that technical advances will narrow the gap between us.

Cost effectiveness is a complex issue, and estimates of cost differ. Regardless of whether true costs are at the low or the high end of published estimates, a broad societal program of fecal occult-blood screening would require either a major shift of current medical resources or a major infusion of new resources — most of which would be consumed by invasive colonic evaluations based on false positive results. Whether this is justified is as much a philosophical issue as an economic one.

Compliance in the Minnesota trial was indeed relatively good, but this cannot be extrapolated to the general public, because the subjects in Minnesota were all recruited from volunteers for the American Cancer Society and other organizations.4 I acknowledged new educational campaigns to improve compliance, but the outcome of such efforts remains to be seen. The literature to date is generally pessimistic.

Bond and his colleagues have made landmark contributions, and I greatly respect their opinions. Their work and that of others make a strong case in favor of occult-blood screening. But there are also compelling contrary arguments — arguments that have not received the attention they deserve.

Jerome B. Simon, M.D.
Queens University, Kingston, ON K7L 5G2, Canada

4 References

1
Church TR, Ederer F, Mandel JS. Fecal occult blood screening in the Minnesota study: sensitivity of the screening test. J Natl Cancer Inst 1997;89:1440-1448
CrossRef | Web of Science | Medline

2
Pignone M, Ransohoff DF. New concepts to understand sensitivity in fecal occult blood testing. Gastroenterology 1997;112:A639-A639 abstract.
Web of Science

3
Simon JB. Fecal occult blood testing: clinical value and limitations. Gastroenterologist 1998;6:66-78
Medline

4
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[Erratum, N Engl J Med 1993;329:672.]
Full Text | Web of Science | Medline

Citing Articles (1)