That obviously depends on the type of cancer. Smokers are way more likely to get lung cancer but for other types of cancer with high mortality rates (prostate, breast, myeloma, lymphoma etc) there is no single attributable 'exposure' that is known to be the cause.
The link to lung cancer is the most high profile (and rightly so -- 80% to 90% of all incidences of long cancer are primarily attributed to smoking), but its carcinogenic effect can be seen all over the body.
Note: Obviously I'm not advising anyone to smoke.
* Eating red meat
* Eating processed meat
* Drinking 3+ glasses of milk daily
* Inadequate levels of vitamin D
These are all well-known to the nutrition community to increase cancer risk.
I think it's overstating it a bit so say those things are "well-known". There are some indications that those things may result in slight increases of risk for certain types of cancer.
I would seriously question the credentials of any PhD nutrition researcher at a decent university that disputed any of those factors.
"The risks of a high-protein diet are even comparable to smoking, the researchers said."[1]
1: http://www.scientificamerican.com/article/diet-high-in-meat-...
http://www.marksdailyapple.com/will-eating-red-meat-kill-you...
Huh? There's some correlation with prostate cancer[1] but that's it.
[1] http://www.pcrm.org/health/health-topics/milk-consumption-an...
Also, "They did not look at tissues from two of the commonest forms of cancer – breast and prostate" this is not an accurate description of 'cancer' just the types of cancer they are studying.
Further, this is focused on types of tissue. Environmental factors that increase risks for all forms of cancer are ignored.
PS: Mathematically, if a very small percentage of the population had much lower risks of cancer it would be almost impossible to detect.
The risk factors are there because we do not do enough to care for each other as individuals in society. The statistics are the tool used for trading liabilities in the applied oncological epidemiology colloquially referred to as health insurance risk modeling.
It's not "Confucian-era Chinese", but google translate gives : 倒霉 which per wiktionary (https://en.wiktionary.org/wiki/倒 and https://en.wiktionary.org/wiki/霉) I don't see anything relating to poverty or impoverishment.
This older report suggested that two thirds of cancer was because of random mutation, rather than genetic predisposition or lifestyle.
Here's an example of the confusing way that older study was reported: http://www.theguardian.com/society/2015/jan/02/two-thirds-ad...
> Two-thirds of adult cancers, say the researchers from the Johns Hopkins Kimmel Cancer Center in the United States, are caused by random mutation in the tissue cells during the ordinary process of stem cell division. In the other third, our genetic inheritance and lifestyles are the main factors.
> The scientists have created a mathematical model which, they say, shows it is wrong to assume that there are such things as “good genes” that may prevent us getting cancer even though we smoke, drink heavily and carry excessive weight.
This new study disputes that, and says that normal random mutation only accounts for 10% to 30% of cancer, and the rest is because of lifestyle or genetic predisposition.
From a public health point of view that's pretty important. Stop smoking, and control how much alcohol you drink, continue to be important public health messages to reduce the incidence of cancer.