The Cancer Journal - Volume 6, Number 6 (November-December 1993)
The question remains open: the method fails
Nowadays, are we all too ready to claim that medicine is a scientific discipline and that it progresses in step with progress in science? At least in developed countries possessing the intellectual framework and the tools of science and the resources to devote to medicine. If, like us, the reader has some doubts as to the truth of this statement, he will read on. If he is upset by doubt or irritated by these thoughts, let him pass over them.
We have chosen to look at three areas of preventive medicine. In each case, it seems that the scientific method has come up against an obstacle, and that common sense has been forgotten. These areas are: the routine examination for blood in stools to detect colo-rectal cancers; the prevention of post-menopausal osteoporosis by long-term treatment with female steroid hormones, and the use of ultrasound scans during all normal pregnancies. The last two subjects belong to the enormous problem of the relationship between the normal and the pathological, which Gershom Zajicek will explore in depth in one of his next editorials (1).
Mandel et al. (2) have shown that systematic screening for occult faecal blood can help to reduce mortality from colo-rectal cancer. Work of this type has been published over the last ten years. Published and contested. One would have hoped that with time, as one study follows another, an increasing rigour in the methods would have rendered the result more and more certain. The work published a few months ago provoked a number of comments; some approving, some criticizing its conclusions. These criticisms are not unfounded. The question remains open.
For the last ten years, a wide-spread belief among general practioners, and perhaps even more so among gynaecologists, has been that the menopause is the beginning of a long period of hormone deficiency in women. They are convinced that all these women should be treated by continuous administration of oestrogens and progestatives, since this is the only way to reduce osteoporosis and its consequences, and to reduce the risk of artheromatic disease and incidents due to cardiac ischaemia. The relative risk of uterine cancer, which would be increased by the use of oestrogens alone, returns to normal if progestatives are included. Felton et al. (3,4) cast a shadow over our optimism by showing that even if these replacement treatments are followed for several years, the women taking them have the same degree of bone mineralization as non treated women after 75 years of age, at the period of life in which the risk of fractures begins to be a real community health problem. Furthermore, it seems that regular physical activity together with a diet rich in calcium can avoid excessive demineralization without resorting to drug treatment. Will the next epidemiological study be able to tell practioners whether their preventive prescriptions are useful or not ? Today the question remains open.
About ten years ago, a consensus was reached on the circumstances indicating an ultrasound scan during pregnancy (5). Let us point out that a normal pregnancy without any identified risk was not considered an indication. This did not, however, prevent the technique from becoming established as an indispensible procedure at least twice during pregnancy, for all women. Now a new conducted study (6) has shown that no advantage can be gained by performing an ultrasound scan without any particular indication; that is, a normal pregnancy in a women without any defined risk. This procedure is not dangerous, but it does not, as was hoped at the beginning, allow us to educe either foetal and neonatal mortality or foetal morbidity. The question remains open.
We have chosen some everyday problems which illustrate the conflict of two attitudes. One is a medical system ready to act, with technical procedures and the prescribing of drugs. The other prefers to wait and see. The first attitude could be called modern, it satisfies public demand and acts according to good sense and scientific reasoning. The second seeks for proof and does not find it. In practice, activism prevails.
We are in an embarrassing position. If medicine is really based on the scientific method, we must resolve this dilemma before the overenthusiasm of one group or the penny-pinching of another inflame the debate.
Of course, the active approach will encourage the sale of drugs, equipment and supplies, and provide employment. Furthermore, it will reassure the healthy that they are well protected by paraphenalia of modern medicine. However, the questions which remain are not only the result of the failure of scientific arguments to stand up to hard selling.
We believe that they are also the result of limits of the scientific method, or at least the part represented by epidemiological analysis and statistics. Do we have an alternative to suggest? At the moment, no. Let us agree, however, that the success of "science" cannot to used to treat these unresolved problems in a conventional way. We suggest that readers who have been seized by doubt return to the work of Paul Feyerabend (7,8).
There are some salutary surprises.
CNRS, BP 8, 94801 Villejuif, France
1. Zajicek G. The normal and the pathological, Cancer J. 1994 sous presse.
2. Mandel J S, Bond J H, Church T R et al. Reducing mortality from colorectal cancer by screening for fecal occult blood, N England J Med. 328, 1365-1371, 1993.
3. Felson D T, Zhang Y, Hannan M T et al. The effect of postmenaopausal estrogen therapy on bone density in elderly women. N England J Med. 329, 1141-1146, 1993.
4. Ettinger B, Grady D. The waning effect of postmenopausal estrogen therapy on osteoporosis. N England J Med 329, 1192-1193, 1993.
5. Office of Medical Applications of Research, National Institute of Health. The use of diagnostic ultrasound imaging during pregnancy. JAMA 252, 669-672, 1984.
6. Ewigman B G, Crane J P, Frigoletto F D et al. The effect of prenatal ultrasound screening on perinatal outcome. N England J Med 329, 821-827, 1993.
7. Feyerabend P. Contre la méthode. Esquisse d'une théorie anarchiste de la connaissance. Le Seuil, Points Sciences, 1988.
8. Feyerabend P. Adieu la raison. Le Seuil, Science Ouverte, 1989.