It becomes apparent when reading through nutrition literature that nearly every food can be shown to be associated with cancer. And this may arouse suspicions in the more cynically minded amongst us. To highlight this phenomenon, a recent paper published in the American Journal of Nutrition identified 50 common foods from a cookbook and searched for academic literature investigating and reporting the ingredient to be associated with cancer1. When the authors found 10 or more studies, they picked the latest 10 for analysis. The results showed that 40 of the 50 studies, 80 % of the total ingredients picked, had articles reporting on cancer risk. Of these 294 studies found by the authors, 191 of them (72 %) contained data that reported them to be associated with cancer. When the authors investigated the use of meta-analyses for these ingredients (large studies pooling data from several previously published studies) they found that 13 out of 50 ingredients (26 %) reported to be associated with cancer.
These results therefore confirm the observation that most foods can be found to be associated with cancer. Many such associations are reported in the mainstream media, and in recent years this has become all too frequent. The main problem with this is one of trust, as it may seem implausible to the layman that so many foods could possibly be associated with cancer. It would not for example be unrealistic to expect that an individual may become desensitised tor confused based on the sheer volume of reports. Central to this is the fact that while science has a duty and a responsibility to report on important findings, including foods that may be associated with cancer, the mainstream media will always take this information and hyperbolise it in order to sell copy. Most people will take this spun and exaggerated information at face value and will desire no further explanation, and good luck to them. However some may feel an explanation for this phenomenon is justified. So why are so many foods associated with cancer?
The main reason so many foods are associated with cancer relates to the semantics of the statement. Firstly an association can be positive or negative. A positive association (or correlation) would imply that as an intake of a particular food increases, the risk of cancer also increases. as might be expect to be found with unhealthy foods. A negative association would imply that as intake of a particular food increases, the risk of cancer decreases, as might be expected with healthy foods. Because an association can include both healthy and unhealthy foods, it is not surprising that so many foods can be shown to be ‘associated with cancer’. Secondly, an association is a link between two variables, but there can be no cause an effect implied in this link. For example, fire engines are associated with fires, but they do not cause fires. The cause of the fire is actually another completely unrelated variable, in scientific terms called a confounding variable. In fact, individual foods may simply be markers for the real cause of cancer, the Western diet.
Evidence is increasingly implicating the modern Western diet in the current epidemic of cancer in the developed nations. However, this is not caused by a single food component, but through years of eating many low quality foods that eventually deplete the immunity of the individual and leave them susceptible to cancer. Individual foods may be associated with cancer simply because they are markers for the Western diet and lifestyle. Pick any food on the Western diet and it will be shown to be associated with cancer because it is part of that diet. The amount of foods that are dangerous enough to individually be a cause of cancer in a lifetime is actually very small. There are some, and these should be avoided. However, the statistical significance of foods associated with cancer, when analysed in a group analysis or a clinical trial, is negligible. This is reflected in the large numbers of individuals required in such trials, as well as the use of statistical tricks such as reporting relative risk rather than absolute risk in the results (here) to increase effects. r
Conversely, it is easy to show in a clinical trial that sugar does not cause cancer and is safe for consumption. If you take a large number of people and feed them sugar for 6 months, it is very unlikely that there would be any difference in cancer rates between those eating a high sugar diet and those eating a low sugar diet. However if you measure a large population of people, you may find a small but significant association between sugar consumption and cancer. This is because the sugar is a marker of the particular diet and lifestyle followed by certain individuals. Those who eat sugary foods are very likely eating a Westernised diet, and therefore will be in poor health, have low immune status, do a high stress job, and other things that increase their risk of cancer. Any foods eaten by such individuals will be associated with cancer for the same reason. When you think of foods in this way, as markers for a diet and lifestyle, it becomes obvious why it is easy to find foods that are associated with cancer.
Focussing on individual foods is therefore problematic because it creates a blinkered impression that does not fit with the nutritional health paradigm. Good nutrition is about seeing the big picture, the diet as an entirety, rather than focussing on one element. That cancer is not present in many indigenous populations before the introduction of Western foods and their associated lifestyle habits is recorded medically. Individual foods from such a diet may be shown to be safe in a 6 month clinical trial, and may only be weakly associated with cancer in large epidemiological trials. But if they are thought of as markers for a low quality diet, the results become more meaningful and the spotlight can be focussed on the real culprit, the diet and lifestyle of the individual. Finding ingredients associated with cancer in recipes taken from Western cookbook is there not surprising, because it is full of Western food ingredients that will be markers of a poor diet and unhealthy lifestyle common in developed nations.
RdB
