Wednesday, April 01, 2009

Meat Intake And Mortality

This study made headlines last week:

Meat Intake and Mortality, A Prospective Study of Over Half a Million People, Archives of Internal Medicine, Mar 23, 2009

It's conclusion:
"Red and processed meat intakes were associated with modest increases in total mortality, cancer mortality, and cardiovascular disease mortality."
Facts:
  • 545,653 (half a million) participants
  • Aged 50 to 71 at baseline
  • Followed for 10 years (1995 - 2005)
  • Meat intake estimated from food frequency questionnaire
  • Questionnaire collected info over the previous 12 months, completed at baseline
  • Meat was grouped into "red" (beef and pork), "white" (poultry and fish), and "processed"
For a taste of what food frequency questionnaires are like, go here (pdf) to see the one used in this study.

The coverage the study received (Study: Lots Of Red Meat Increases Mortality Risk, Daily Red Meat Raises Chances Of Dying Early) was a little too enthusiastic, I thought. This study provided evidence that red meat and processed meat each increase risk for mortality. It suggests a hypothesis. It cannot, by nature of its methods, claim that high meat intake results in earlier death.

It does, however, add quite a big data-dump (half-a-million participants, 10 years of follow up) to the growing body of evidence supporting this association.
________
Comments:

Typing

This was interesting:

"Subjects who consumed more red meat tended to be:
  • Married
  • More likely of non-Hispanic white ethnicity
  • More likely a current smoker
  • Have a higher body mass index
  • Have a higher daily intake of energy, total fat, and saturated fat
  • Tended to have lower education
  • Tended to have lower physical activity levels
  • Tended to have lower fruit, vegetable, fiber, and vitamin supplement intakes."
Of course, all those confounders were adjusted for in the final analysis. Red meat intake, independent of those variables, was still associated with mortality.

White Meat Protective?

This was notable:
"When comparing the highest with the lowest quintile of white meat intake, there was an inverse association [except for CVD mortality in men] for total mortality and cancer mortality, as well as all other deaths for both men and women." [Emphasis mine.]
So, there was a presumed protective effect from eating white meat (rather, substituting white for red) - except for men. The more white meat men ate, the greater their risk for heart disease. (White meat included chicken, turkey, and fish.)

The association was really small though, so any protective effect of substituting white for red, if it existed at all, may also be small.

More Fuel

The associations were consistently dose dependant. The more red meat and processed meat eaten (as you moved up the quintiles), the greater the risk for mortality. This boosts the credibility of the association.

Population Attributable Risks (Or How Many Deaths Could Have Been Prevented)

The authors calculated:
"For overall mortality, 11% of deaths in men and 16% of deaths in women could be prevented if people decreased their red meat consumption to the level of intake in the first quintile." (Women could experience an even greater 21% decrease in mortality from heart disease by cutting back.)

What Is It About Meat That Could Raise Mortality Risk?

Possible mechanisms:
  • Carcinogens in cooked meat (heterocyclic amines, polycyclic aromatic hydrocarbons)
  • Carcinogenic N-nitroso compounds (in processed meat)
  • Iron (increases oxidation damage)
  • Organic pollutants (pesticides, dioxin, PCB) bioaccumulate in animal fat
  • Saturated fat (linked to breast and colon cancer)

What Did Meat Groups Say?

The American Meat Institute, the National Pork Board, and the National Cattlemen's Beef Association dismissed the findings.
________

24 comments:

Jennywenny said...

This was an interesting study. It does seem that the american diet is massively high in meat compared to what is really required.

I listened to the food program this week on the radio and thought you might be interested in what they had to say.
http://www.bbc.co.uk/programmes/b006qnx3
The omega 6/omega 3 thing isnt new, but I was really interested in the fact they mentioned that omega 6 oils come primarily in nuts and seeds and are a wintertime food, slowing our metabolism etc.

BizBuzz said...

WOW, that was totally interesting - thanks for posting all those facts about this. I am gonna share this with my group.

caulfieldkid said...

Hey Bix,

Do you have any idea as to what 9.3g/1000 kcal looks like (referenced on the Q1 column for the red meat line)? I'm just curious as to how much we are talking about here in more common terms.

As you pointed out, correlation doesn't always equate to causation. From looking at the chart, the nutritional life style of the participants degraded, over all, as you moved towards Q5. Red meat may certainly be a part of the problem, but it seems to me there is more going on here. I wonder if there are other factors that would correlate (urban vs. rural; white collar vs. blue; income etc.).

One other observation: The study seemed to focus heavily on whites (not that it means anything).

Bix said...

Nicely done program, Jenny. Thank you for that.

For a few years now I've been saying that the best way to increase your (relative) amount of omega-3 was to reduce your intake of omega-6. Both are polyunsaturates, not as stable as the mono's and saturateds, and have their own problems.

First time I heard about the seeds slowing metabolism. I'll have to read up on that.

Diet analysis programs have a pretty large margin of error. If you're eating grass-fed beef and chickens that pecked, you're probably getting a higher N-3:N-6 ratio than what is typically reflected in those programs.

Bix said...

shaun ...
That would be, what, 18.6g/2000 or a little more than half an ounce/day of red meat. ... That would look like one fast food burger (quarter pounder) a week. The rest of your meat for the week would be white (chicken/fish).

"One other observation: The study seemed to focus heavily on whites (not that it means anything)."

Good point. The characteristics of a population matter. Technically, you can only generalize (apply) your results to a population that looks like the one you studied.

Bix said...

Biz, I never heard of the hCG diet. That's interesting.

Bix said...

shaun, I hear you about "more going on here." There always is.

The hazard ratios were low, thus the authors' statement of "modest increases."

When an epi study doesn't result in stronger associations (but statistically valid associations) it could mean unmeasured covariates were confounding the association.

Those unmeasured covariates could send the association either way:

- Towards supporting the null hypothesis (that there is no association between red meat and mortality), or,
- Towards rejecting the null hypothesis (that there is a clear association between red meat consumption and earlier death)

If you say that the strength of the association was too low, and that some other factor was influencing death besides meat (e.g. as you said, urban/rural, etc.), then you are, in effect, recognizing the alternative argument, that some other factor was interfering with a bona-fide meat-death correlation (e.g. in food questionnaires people tend to underreport rather than overreport. If people didn't come clean about how much meat they ate, that could result in an observed weaker association than is actual.)

Unknown said...

i know a couple of south east asian families who has goat meat - 3 times a day (if they could ) and heavy on nuts. the families are as healthy they can be -- fat is an overstatement -- they also have a lot of beans - but not greens or fruits as much in their diet. their men are heavy smokers and drinkers -- i don't know what is the secret of their diet --- and they are not physically active either! but boy they have good health ..

speaking about omega 3 -- i have observed chinese families at whole foods who have told me - stock up on flax seeds for econommical omega 3 regularly -- ( they were so generous with the info )

again i think it is the regional foods that give you good health.

americans have a european way of diet --may be it is time to understand the land and the soil and the lifestyle and see what grows in this soil and make a fresh start -- my 2 cents.

( i don't think what i have written here is as much as with the study ...still i thought it might be interesting..

Matt said...

I'm surprised to see that white meat and fish still caused higher rates of heart disease in men. I think I need more clarification. Is this linked directly to the fact that they were eating meat or were they also eating few vegetables?

Anonymous said...

Absolutely love your blog! I am a first time visitor and was just curious about your take on the meat/ health debate. I noticed that you posted a fairly vegetarian bean and barley recipe (looks great, by the way). However, you seem to be unimpressed by the research suggesting that meat, especially red meat, adversely affects health. Based on this and other studies, what type of role do you think that meat has in a healthy diet? Thanks for your thoughts!

Bix said...

Hi Anonymous. That was nice of you. Thank you.

So, where do I stand on meat? In the scheme of things, I don't think my thoughts matter. But I do have thoughts, and since you asked ... (See my latest post.)

So, my turn :)
What are your thoughts on the role of meat in diet?

Angela and Melinda said...

To Anrosh--goat meat three times a day--thats a lotta meat!!!! Even if you can afford it. How old are the oldest family members?

To Bix--What's a hazard ratio? And thanks for another really illuminating article on meat. I think your general conclusions are spot-on.

Unknown said...

Dr Mel - meat is not the main course for their meal- it is always a side. There would be rice/rotis made out of whole wheat (generally made at home ) or pitas. And there would be lentils or beans too.

Bix said...

Melinda, hazard ratio (HR) is a statistic used to interpret data. It's often used in epidemiological studies because you can't do a relative risk (RR) because you can't get an exact incidence. It's like an estimate of RR; it's not time dependant though.

To interpret:
An HR of 1.36 means that people who ate the amount of red meat in a defined quintile (say the top) had a 36% higher risk of death than people who ate the amount of red meat in a comparative defined quintile (say the bottom) - at any given point in time.

Bix said...

I also think red meat three times a day is a lot. I guess if other sources of food are limited, that's where you'd get the bulk of your calories.

I know that people in Mongolia get a big chuck of their calories from animal foods, but growing is difficult there. There's also a relatively high rate of malnutrition.

Angela and Melinda said...

Anrosh and Bix, thanks for the response and explanation!

Anonymous said...

Bix,
You inquired about my stance on meat. This is one that I have really struggled with over the past decade.

I am a nutrition professor and former vegetarian (pesco-lacto-ovo). When I became pregnant with my first child, I was concerned about some of the research suggesting that iron and zinc deficiencies may impair fetal and infant development. Of course, a properly planned vegetarian diet can provide these nutrients. But I didn't believe that the benefits of a meat-free diet outweighed the risks for my child's development.

Even during my vegetarian years, I never saw any evidence that a vegetarian diet was healtheir than plant-based diet(e.g., low meat). And of course there are concerns about low EPA/ DHA, B12, iron, and zinc in vegans.

I guess I have come to the conclusion that small amounts of meat can fit into a healthy diet if one chooses to eat them. I agree with you about B12. If it weren't for that one little nutrient, I would say that we don't really have a physiological need for meat.

Bix said...

Anonymous, Thank you for visiting, for asking me my thoughts, and for sharing your own.

You have me thinking about what it must be like to be, say, a devout vegan who is compelled to instruct on the benefits of meat-eating. Or someone with a strong meat ethic compelled to instruct on the benefits of vegetarianism. Which has me thinking about what it must be like to be someone with strong religious beliefs who is compelled to teach, say, evolution. Or an atheist who is compelled to present Creationism in science class. Okay, I'll stop here. I'm left, after that thinking, with a greater respect for teachers.

Angela and Melinda said...

Question for Anonymous and Bix: Anonymous, you said you were a pesco/lacto/ovo vegetarian, but that you were concerned you couldn't get enough iron & zinc to support a fetus. Then you mention EPA/DHA and B12 issues w/ vegans. My question (actually a couple) is, do fish lack iron & zinc? Do dairy products like aged cheeses (from red-meat animals) or eggs lack those minerals?

From the way you phrased it, I assume that one does find B12 in fish (and of course the oils). Is that correct? I ask b/c I'm a vegetarian who generally does not eat fish (only very infrequently and only small fish like sardines, etc.). However, the study you cite, Bix, about long healthy lives, combined w/ what you and Anonymous say in these meat comments, are making me wonder if I should include a bit more fish (or meat), tiny bits, maybe a couple times a week. Just wondered if I were interpreting your comments correctly re missing nutrients in vegan/veggie diets.

Bix said...

I'll take a stab, Melinda.

B12, like all the B's, is water soluble. So you won't find it naturally in fat or oil (like fish oil). It's one nutrient that's difficult to get unless you eat animal foods (fish or livestock), or take a supplement, or eat supplemented foods liked breakfast cereals.

Iron you can find in lots of foods, plant and animal (including seafood). The animal sources often come in a form (heme) that is better absorbed. Dairy foods contain practically no iron, and may interfere with absorption of iron in other foods. Iron supplementation or consumption of foods high in iron isn't a good idea when iron loss (blood loss) isn't occurring.

Zinc is found more in animal foods (including seafood) than plant foods. It's more bioavailable in its animal form, like iron. As with many nutrients, however, the less you eat, the more you absorb. This is one way that people who consume just a few hundred milligrams of calcium a day can have good calcium status when measured.

Angela and Melinda said...

*Very* interesting, especially the part re getting more from taking less--kind of zen! Thanks so much for the info!

Matt said...

Hi! Stumbled upon your blog through blogger (clicking 'next blog' which I rarely do!).

I have done a lot of reading on the paleo side of things ... we evolved to eat meat, veg, omega6-omega-3 ratio of 1/1, to eat cholesterol, to not regard saturated fat as troublesome, but to regard processed omega-6 fats as such. To minimize carbohydrates and avoid wheat, but realize that vegetables and occasional fruits are beneficial. Curious to learn your thoughts on a paleo plan.

Bix said...

Hi Matt,

Thank you for visiting!

I'd go along with most everything you said. That's how we ate at some point in our past. Since it brought us from there to here, it must have been doing something right.

Your paleo diet question is a good one. I think I'll put some of my thoughts in a post and let others comment too.

Angela and Melinda said...

And there's this to consider too: http://www.theonion.com/articles/study-red-meat-takes-years-off-of-cows-life,27954/?utm_source=Facebook&utm_medium=SocialMarketing&utm_campaign=standard-post:headline:default