[Help] [Aide] [Up]

The Cancer Journal vol. 6, num.3 (May-June 1993)


A disappointing balance sheet in Sweden

Large epidemiological studies of cancer are rare. Very few countries have a well equipped statistical service which can supply data over a time period sufficient to allow significant conclusions to be drawn about long-term trends. However, this is the only type of result which comes close to the truth about cancers: risks, incidence, prevention, the effects of treatment and the impact of innovations which attract considerable and often misleading attention from the media.

The small number of rigorous studies which have come to our knowledge carry much more weight for the evaluation of the situation than the thousands of publications of experimental results which make up the more visible part of the literature in cancerology.

We are therefore consider ourselves bound to analyse each of these rare glimpses of a reality which is escaping us. In this editorial we would like to comment on one such case; the work of Adami et al. (Lancet 341, 773-777, 1993). The authors have used the Swedish National Cancer Registry, which includes 837,085 cases from 1958 to 1987; 418,624 in women and 418,461 in men. Over these thirty years, the incidence of cancer increased by 30% for women and by 55% for men. The increase was more marked for the age ranges above 50 years old; and also for subjects between 20 and 30. The tendency towards increased incidence was less apparent for the 30 to 40 year-old age group, but was much greater in the older age groups, especially for men.

Adami et al. divided the population into cohorts, which brought out the fact that the relative risk rose from 1.0 in women born between 1873 and 1882 to 1.7 for women born after 1920. This increase in the relative risk passed through a series of plateaus. As far as men were concerned, the cohort study showed that the relative risk increased steadily, and was three times as large for men born int he 1950s than for men born in the 1880s.

Although this trend towards increased risk was greatest for cancers linked to tobacco use, it was also present for cancers not believed to be associated with smoking.

In Sweden, the increased incidence could be set against an improvement in survival to explain the relatively unchanged mortality rate. For younger subjects, the decreased mortality noted could be almost completely attributed to improvements in therapy. It is unlikely that the increased incidence could be due to more efficient detection in this age group.

On the other hand, efforts made towards early diagnosis have certainly played a role in the over-sixty age group, and can at least in part account for the increased incidence; for example, prostate cancers which hitherto remained undetected. Systematic screening seems to have helped to reduce the incidence of cervical cancer, without having any appreciable effect on the incidence of any other type of cancer.

The authors consider the rising incidence from one cohort group to the next since the end of the 19th century as indicating increased exposure of the population to carcinogens.

When these results are taken together with the high cancer mortality rates in the United States and Germany, one cannot help questioning the ability of modern medicine to cope with the growing pressure of cancer in adults, despite the deployment of considerable technical resources. It is clear that, in this field, technical excellence is not enough and that something else is needed if progress, so often announced, is to become a reality.

Jean-Claude Salomon
To contact the author...Click here Thank you.