Counting and statistics are necessary but not sufficient methods in search of causes, and this still holds true when digging deeper into details.
Scenario 1: Over the past century, the number of storks has diminished in all industrialized countries, and so have birth rates. Today, in countries where storks are abundant, for instance in Africa where they stay in winter, birth rates are also high. There is a clear, highly significant positive correlation between the frequencies of storks and babies. There is also a theory about storks bringing children. Do storks cause births?
Scenario 2: In human populations, under certain conditions, an increase in both fat cells and cancer cells has been observed - not in general but only in some selected studies. This positive correlation between fat cells and cancer cells is weaker than in the stork and baby case. There are also some theories about body fat cells producing substances that may promote tumor growth. Do body fat cells cause cancer?
Scenario 3: In human populations where the number of smoked cigarettes has increased, the number of cancer cells, mainly in the lung, also has increased. There is a clear, highly significant positive correlation between the frequencies of smoked cigarettes and lung cancers. There is also a theory why and how harmful chemicals in tobacco smoke cause damage in lung cells, turning them into cancer cells. This theory has been tested in a huge number of scientific trials, with positive and convincing results. Does tobacco smoking cause cancer?
Cause or not cause, that is the question
The difference in the three scenarios above is not in the correlations but in facts that must be dug out, looking deeper into details. Don't get me wrong, correlations are important, without correlation there cannot be a cause. But correlations, while necessary, are far from being sufficient, as the first scenario shows. The reason storks do not cause births is because this cannot work in detail.
In scenario 3, the cause is generally accepted and can be taken as a proven fact because we know many details about how things work between toxic chemicals and body cells.
Scenario 2 is somewhere in between. There is some amount of correlation, not without doubt, but let's accept it as given for the moment. Some researchers claim having found evidence for mechanisms how body fat cells cause damage that may turn healthy cells to cancer cells. Now let's take a look at these details and see how convincing they are.
Substances from body fat cells
The main theory of how body fat should cause cancer relies on inflammation. In chapter 2, the authors of the food and cancer report tell us how:
"Obesity is characterised by a low-grade chronic inflammatory state, with up to 40 per cent of fat tissue comprising macrophages. The adipocyte (fat cell) produces pro-inflammatory factors, and obese individuals have elevated concentrations of circulating tumour necrosis factor (TNF)-alpha, interleukin (IL)-6, and C-reactive protein, compared with lean people, as well as of leptin, which also functions as an inflammatory cytokine. Such chronic inflammation can promote cancer development."In short, body fat cells produce a number of substances that go together with inflammation, and inflammation is thought to damage body cells, turning them into cancer cells.
But exactly the first of the mentioned substances, tumor necrosis factor alpha, does not cause cancer but helps preventing it. Necrosis means death, and this factor makes tumor cells die. Other substances may promote cancer, but this one prevents cancer. Thus, the action of all these substances from body fat may be a zero sum game when it comes to cancer.
And when scientists are digging more into details, they may find stork and baby cases even in a deeper level. I'll deal with this question later. Stay tuned.
Photo credit: flickr.com/photos/millan/95188464/