A ham sandwich a week ... dead in 6 months?

January 20, 2019 3:50 PM

Help me interpret this dietary warning about processed meat.

“We see a 4 percent increase in the risk of cancer even at 15 grams a day, which is a single slice of ham on a sandwich,” said Dr. Nigel Brockton, director of research for the American Institute for Cancer Research. -- New York Times, Dec 14 2018

I don't question this conclusion from a review of studies. But what does it mean?
For example:
● Eat a ham sandwich at lunch and your risk of keeling over dead before dinner rises 4%.
● Do it again tomorrow and you're 8% more likely to check out that day.
● Have a regular lunch of ham sandwiches and whoa! -- you're 100% dead by month's end.
● Or some other metric.

I understand risk, or thought I did. And no, I don't seriously think any of these interpretations is correct. But what is?

posted by LonnieK to Health & Fitness (12 answers total)

• Eating 15 grams of processed meat every day raises your risk of cancer by 4%.

posted by zamboni at 3:56 PM on January 20, 2019 [8 favorites]

Frankly that kind of statement is utterly useless (worse than useless actually, because it spreads fear) for exactly the reasons you have struck upon. It also doesn't clarify what the baseline rate of cancer is, and it doesn't say whether it's an even increase across all cancer types, or just certain cancers, or whether it increases all cancer risk but some types
more than others. Not all cancers are equally bad, so that's pretty important.

When I see a statistic like that, I just ignore it. It's just noise. It sounds sciencey, but it means nothing—there's probably some truth down at the core somewhere, but without better information I just don't consider it actionable.
posted by Anticipation Of A New Lover's Arrival, The at 3:56 PM on January 20, 2019 [9 favorites]

I don't know the actual number, but it's something like this:

A normal person has an 11.5% chance of developing cancer in their lifetime.
If you eat 15 grams of ham every day, you raise that to 11.96% (11.5 x 1.04)
posted by Hatashran at 3:57 PM on January 20, 2019 [38 favorites]

From the study the article is citing:

Conclusions
The current evidence from prospective studies supports limiting the amount of red and processed meat in the high consumers for colorectal cancer prevention. Primary prevention of colorectal cancer should emphasize modification of multiple diet and lifestyle factors.

So yeah, the headline of the NYT piece doesn't match the conclusion of the meta-analysis it purports to report on.
posted by mandolin conspiracy at 3:58 PM on January 20, 2019 [5 favorites]

It also doesn't say whether it's an increase in lifetime cancer risk or risk over a certain period of time, and it doesn't say over what period of time you need to eat a slice a day to get that 4% increase. There's just so much missing information there that the only sensible thing to do is dismiss it until you hear something more specific.
posted by Anticipation Of A New Lover's Arrival, The at 3:58 PM on January 20, 2019 [3 favorites]

Yes after eating ham every day, risk goes up by 4%. And my guess is that the baseline rate is multiplied by 1.04, not that 4% is added to the baseline rate. Like, from 10% to 10.4%, not 10% to 14%.
posted by supercres at 3:59 PM on January 20, 2019 [7 favorites]

On nonpreview, what Hatashran said.
posted by supercres at 3:59 PM on January 20, 2019

It also doesn't clarify what the baseline rate of cancer is, and it doesn't say whether it's an even increase across all cancer types, or just certain cancers, or whether it increases all cancer risk but some types
more than others. Not all cancers are equally bad, so that's pretty important.


If you look at the article, it's specifically about colorectal cancer.
posted by FencingGal at 4:12 PM on January 20, 2019 [2 favorites]

The risk for getting colon cancer for the average individual is around 5%.

Eating 15g of processed meat a day increases this to 5 x 1.04 = 5.2%. So eating processed meat increases by an additive amount only .2% on average.

Eating 10g of fiber a day decreases the average risk to 5 x .90 = 4.5%

Smoking increases the average risk to 5 x 1.2 = 6%

A multitude of other choices - cause changes in risk at about this level. If eating processed meat increased risk from 5% to 9% that would be reported as 80% increase in risk - this would also likely result in the removal of processed meat from the market as a cancer causing agent. The complicating factor here is the fact that we're looking at percentage changes in percentages - a process that can unfortunately be used to exaggerate or minimize changes.

Colon cancer is easily detected and very survivable if detected early. If you are at a high risk, or getting older make sure you get a routine colonoscopy to check.
posted by NoDef at 7:12 PM on January 20, 2019 [10 favorites]

What it means is that scientists don't get funding for their next study unless they find a way to make the results of their last study sound Very Important.
posted by tobascodagama at 9:05 PM on January 20, 2019 [2 favorites]

Not knocking the original meta-analysis they are referring to, BUT. The American Institute for Cancer Research is a (poorly rated) charity focused on diet and weight management as relevant to cancer risk. Don’t confuse it with the national governmental organization, the National Cancer Institute.
posted by Knowyournuts at 11:06 PM on January 20, 2019 [5 favorites]

I don't know the actual number, but it's something like this:

FYI it's about 40-50% depending on data source and method. Which, yes, is a fucking lot - but it should be noted that a) epidemiologists like to say that cancer is a disease of aging (e.g., the older you are, the greater your risk), and b) these numbers are based off of the current population, which skews older and frankly unhealthier - habits from 50 years ago (around smoking, diet, exercise, for example) factor in to this.
posted by entropone at 6:32 AM on January 21, 2019 [2 favorites]