7 ms·
The author fails to meaningfully contextualise the risk, giving what I think is a completely misleading account. According to the best available evidence, eati
by jdietrich 9y ago
The author fails to meaningfully contextualise the risk, giving what I think is a completely misleading account.
According to the best available evidence, eating 50g per day of processed meat increases the relative risk of developing colorectal cancer by 18%. Given the prevalence and prognosis of colorectal cancer, this equates to about one premature death per 200 people as a result of eating processed meat.
Is processed meat somewhat bad for your health? Almost certainly. Is your daily bacon sandwich worth a 0.5% chance of a premature death? That's entirely your decision. I'm a vegetarian for ethical reasons, but that level of risk wouldn't faze me in the least if I really liked bacon. Factors like obesity, inactivity, smoking and alcohol use dwarf the risks posed by processed meat.
Constant fretting over small risks posed by our diet may be distracting us from clear messaging on big risks. I've heard countless people say "everything gives you cancer, so why worry about it?" or "scientists can't make their mind up about what you're supposed to eat, so I just ignore them". It isn't news that smoking, being obese or drinking too much alcohol are incredibly bad for your health, but it's a message that deserves repeating. I fear that those messages are being drowned out by a constant trickle of stories about "superfoods" and cancer risks.
The author of this article has previously made exactly this point in an article criticising the "clean eating" fad:
"When mainstream diets start to sicken people, it is unsurprising that many of us should seek other ways of eating to keep ourselves safe from harm. Our collective anxiety around diet was exacerbated by a general impression that mainstream scientific advice on diet – inflated by newspaper headlines – could not be trusted. First these so-called experts tell us to avoid fat, then sugar, and all the while people get less and less healthy. What will these “experts” say next, and why should we believe them?"
https://www.theguardian.com/lifeandstyle/2017/aug/11/why-we-fell-for-clean-eating https://www.theguardian.com/lifeandstyle/2017/aug/11/why-we-...
- techman9 9y agoThe concrete numbers about absolute risk are useful. Thanks.
- coldtea 9y ago>According to the best available evidence, eating 50g per day of processed meat increases the relative risk of developing colorectal cancer by 18%. Given the prevalence and prognosis of colorectal cancer, this equates to about one premature death per 200 people as a result of eating processed meat. That is a huge number.
- jdietrich 9y ago>That is a huge number. Compared to what?
- coldtea 9y agoCompared to the lifetime odds of almost all other mortality risks: https://www.iii.org/fact-statistic/facts-statistics-mortality-risk https://www.iii.org/fact-statistic/facts-statistics-mortalit... Not to mention just in itself -- opposed to 0 people dying prematurely because of that.
- 13of40 9y agohttps://amp-businessinsider-com.cdn.ampproject.org/i/s/amp.businessinsider.com/images/5a860c28d0307219008b45d4-1136-1321.png https://amp-businessinsider-com.cdn.ampproject.org/i/s/amp.b... Take your pick. It's basically worse than gun violence but better than alcohol.
- fjsolwmv 9y agoIt's premature, but at an age where non-meat-cancered people are likely to get nipped by something else anyway
- slx26 9y agoSorry, how did you get to that 0.5%? I'm not sure it's correct. Haven't looked at it enough, but I was thinking that if relative risk of colorectal cancer increases by 18%, if there are ~40 cases per 100.000 people, then it's going from 0.04% to ~0.05%. Didn't even read the article, if anyone can run some numbers and clarify? (genuinely curious)
- colin_mccabe 9y agoYes, you did it right. The parent comment was unfortunately incorrect. To me this is an absolute non-issue. Even if the research is free of bias and error (debatable), it's a tiny chance of cancer late in life. Spend your time worrying about almost anything else (diet, exercise, avoiding drugs), etc. and it will be orders of magnitude more important.
- jdietrich 9y agoThere are 40 cases per 100,000 people in any given year. Your chance of developing colorectal cancer at any point in your life is about 4.3%, with a survival rate at five years of about 65%. An 18% increase in risk of developing colorectal cancer equates to about a 0.5% chance of dying prematurely. Some people will survive beyond five years but die of cancer anyway; I didn't bother factoring this in, partly because I'm lazy, partly because I couldn't find relevant data and partly because the median age at diagnosis for colorectal cancer is 67. https://seer.cancer.gov/statfacts/html/colorect.html https://seer.cancer.gov/statfacts/html/colorect.html
- fjsolwmv 9y agoDying at 70 is barely premature. general physical health (obesity and strength) and accidents are far more an influence.
- riku_iki 9y ago> 0.5% chance of dying prematurely. > the median age at diagnosis for colorectal cancer is 67. So statistically speaking eating 50g of processed meat a day reduces your life expectancy by 3.2 days: (78 years life expectancy in US - 67 median age of getting cancer - 1 year of dying from cancer) * 0.005 chance of receiving cancer * 0.18 increase of cancer probability * 365 days.
- bayonetz 9y agoOn a related note, as a new parent, I’ve been inundated with advice where no sense of absolute risk or prior probability is given. For example, stuff like “don’t sleep baby on stomach because of increased SIDS risk” or “not inducing at 41 weeks increases chance of stillbirth”. It’s like what am I supposed to do with that? Without additional quantification, it’s hard to do any kind of cost-benefit analysis. Things like convenience, sustainability, and financial burden matter too and you need to be able to find the right trade off. Even vague casual quantifications like “the chances are super duper low” would be sufficient in most cases to take action upon.
- fjsolwmv 9y agoWhat value of numbers would change your mind about the trade-off between sleep quality and risk of sudden death? SIDS incidence is extremely rare, s you check by counting how many babies you've known and how many died.
- js2 9y agoConstant fretting over small risks posed by our diet may be distracting us from clear messaging on big risks. Apropos enough, this week’s episode of The People’s Pharamacy is about exactly this: https://www.peoplespharmacy.com/2018/03/01/show-1112-how-and-why-to-eat-sinfully/ https://www.peoplespharmacy.com/2018/03/01/show-1112-how-and...