CANCER: EPIDEMIOLOGY – HOW CONCLUSIVE (part 2)The central difficulty in interpreting the results of epidemiological studies lies in distinguishing whether a link that appears to exist between a risk factor and a cancer is truly a causal one. For everybody except epidemiologists, this problem requires Quite a lot of explanation. The studies may show that a particular group within the population is more likely to get a particular kind of cancer. Let us say, for example, that the members of a particular occupation are more likely to get stomach cancer. Some particular feature of that occupation attracts attention. Let us suppose that the occupation is associated with extensive foreign travel. We know that the members of that occupation are more likely to have stomach cancer than the general population, so we might look for a link between the number of miles flown in an aeroplane and the incidence of stomach cancer. We will find that there is an association. These people travel abroad much more than people whose jobs keep them at home. Does air travel cause stomach cancer? The link is there from this epidemiological study but it may not mean that air travel causes cancer. There may be something else about this population which explains that high incidence of stomach cancer. Could it be that their foreign travel takes them to parts of the world where the diet is very unusual?Perhaps they are eating something abroad which is associated with stomach cancer. Perhaps, on the other hand, these highflying foreign travellers are very well paid and are therefore living lavish lifestyles, eating lots of rich food and drinking lots of alcohol. Is it this that is the causative factor for their stomach cancer? The point we are making is simply that the establishment of a link between a cancer and a particular feature of someone’s life can never prove in itself that that link is responsible. If the link is a very strong one and if it seems to be sensible in biological terms (perhaps diet is more likely than air travel to be a factor in stomach cancer), then it can be an important due, but it doesn’t demonstrate proof of the cause, and much more work is necessary before we can assert with confidence that there is a causative link.We should emphasize that we have used the example of air travel and stomach cancer simply as an illustration of the need to be careful in interpreting epidemiological evidence. There is no need for members of the travel industry to go rushing off to sec their doctors!Some of the problems which we encounter in trying to prove causal links loom very large at the present time. The best example is the possible link between dietary fat and breast cancer. We will discuss this topic in detail in a later chapter but one point can be made here. Breast cancer is much commoner in countries where women eat a lot of animal fat. This might mean that animal fat causes breast cancer. However, the countries in which we eat a lot of animal fat are also the wealthy countries of the world and they have many other distinctive features. Work patterns are different and leisure activities are different. In wealthy countries women are much more likely to own a television set. It follows therefore that those countries where television ownership is high have a high incidence of breast cancer. Television doesn’t cause breast cancer. The link exists because of the association with high standards of living. A high intake of animal fat may or may not rum out to be a cause of breast cancer; unravelling this factor from other factors which might contribute to the development of breast cancer is a major target of epidemiological studies at present.*20\194\4*

Some cancers start very close to the surface and can be seen and felt as a lump under the skin by the patient and the doctor. Common ones are cancers of the breast, lymph nodes (primary or secondary), and testicle. A lump under the skin can be the first indication, of other cancers also, such as ones starting in fatty tissues, muscle, bone etc. Of course, there are many other possible causes besides cancer for lumps under the skin.

*60/40/1*