Can we PLEASE look at what they included in the study when they do this stuff. Did you notice they included "We used nationally-representative cross-sectional survey data to analyze consumption of unprocessed red meat; processed meat; and total red and processed meat."
In the study this includes "processed meat (defined as all meats that have been salted, cured, fermented, smoked, or otherwise processed for preservation and flavor enhancement)."
Flavor enchancement? "Additionally, flavor enhancers such as monosodium glutamate (MSG) and disodium 5'-ribonucleotides are used to amplify the savory taste of umami.
Hydrolyzed vegetable protein (HVP) is another flavor enhancer that brings out the savory, meaty flavor in processed meats, making it popular in vegetarian and vegan food products."
PROCESSED MEAT includes "Processed red meat refers to red meat that has been preserved through methods such as smoking, curing, salting, or adding chemical preservatives. This includes meats like ham, bacon, salami, and some sausages."
Adding chemical preservatives? Defined as "Processed meat often contains various chemical enhancements to improve its taste, texture, and shelf life. Common preservatives include sodium nitrates and sodium nitrites, which can increase the risk of heart disease and diabetes according to Mayo Clinic licensed dietician Kathrine Zeratsky.
Other additives include hydrolyzed protein, modified food starch, and monosodium glutamate (MSG).
These additives enhance flavor and texture but may introduce unnatural substances into the human body, potentially leading to harmful side effects and health issues.
Additionally, processed meats can contain artificial colorants, flavorings, and texturants, which are used to improve appearance and taste."
Good point. Ambiguity as to what participants were actually eating is one reason the approach to measurement in this study was a hot mess. There's nothing in the study itself or the supplementary materials that adequately addresses your point. The food survey questions weren't (and probably couldn't be) specific enough.
I don't blame the researchers per se for the problem, because it's inherent to large observational studies - i.e., there's no way to accurately record what each of tens or hundreds of thousands of people eat each day over a period of years.
Practically speaking, it seems reasonable to avoid eating too much processed meat, on the grounds that something in it (though we can't say what) increases risk of health problems if we eat too much of it (though we can't be sure what the safe maximum would be).
The other thing is when people say “RED MEAT IS BAD FOR YOU” are they talking about the re meat sold at Wamart that is injected with “water and chemicals to increase its weight and improve its appearance. These chemicals can include salt, phosphates, antioxidants, and flavorings.”
OR
are they talking about your local grass-fed, grass finished farmer?
Yeah... individual, large-n observational studies are rarely set up in a way that allows for such fine distinctions (e.g., processed red meats with high vs. low sodium content). Rather it's the totality of evidence (observational, experimental, petri dish, etc.) which suggests that red meat and salt, for instance, each increase health risks beyond certain intake levels.
Interesting until YOUR political biases ruined your credibility. You presented small bits of those political examples, much like the investigators didn’t report important data, or acknowledge it was missing. I’ll read the study myself.
I appreciate your point, but I think that my credibility should only be judged on the basis of how accurately and insightfully I reviewed the study, because: (a) I openly acknowledged my political biases, and (b) I presented evidence for my politically partisan complaints.
On occasion I air partisan complaints without evidence, but 100% of the time those are brief, snarky comments, and they have no bearing on my treatment of the data.
I wouldn't claim to be completely impartial - nobody is - but I do try my best to present nonpartisan discussions of study findings, and in some posts, like a recent one on butter and mortality, I end up criticizing elements of what might be called a liberal perspective.
As you point out so well, diet studies, especially those relying upon self-reporting, are fraught with errors. Meta-analyses suggesting trends, rather than single studies giving percentages, are far more persuasive, but don't seem to inspire as many headlines.
It's hard to get people interested in study methods. We're busy; we just want to know the takeaway.
I wish high school curricular standards made more room for supporting data literacy, because even kids at that age can appreciate the difference between one crappy observational study vs. a well-done meta-analysis (even if the stats are beyond what's covered in AP Statistics).
Countries such as Finland begin teaching critical thinking skills in elementary school. Just another area where the U.S. lags behind other developed nations.
Nice work and agree with your assessment of what seems to be an arbitrary and capricious criteria for the processed groups with rather low bars. You mentioned the "Informational siloing also seems to be occurring." This reminded me of the new proposed dietary guidelines that emphasise plant based over meat I shared earlier this month. Seeing the same thing, of course, thin s might have changed since then. Guess there is a global plot to take away your burger. Anxieties at the Petri Dish - by KB's FROM THE PETRI DISH
Next posting is going to look at a few things on diets, the gut biome and brain/ling/heart axis. Suspect there is an underlining commonality to this and other medical conditions.
It's sad to see that issues like diet-health relationships are so deeply politicized. I understand that people are going to follow party lines when debating trade-offs between the economic benefits and the environmental damage created by the cattle industry, for instance, but it's a shame we can't just set politics aside when looking at the impact of beef consumption on health and let the data lead wherever it may.
Dementia or Alzheimers is an interesting endpoint, too, because cured meats like hot dogs and salami and pepperoni are fairly significant hazards for various cancers (1.1 - 1.4 HR), and of course red meat is also a significant CVD hazard (1.1 - 1.3 HR depending on unprocessed vs processed and consumption), likely due to saturated fat (and nitrites and nitrates in the processed / cured ones).
There's not necessarily any direct relations between those endpoints and dementia, but it definitely feels like there's probably additional dementia risk if you have CVD or higher cholesterol (and correspondingly higher inflammation), or if your body is under additional immunological and inflammatory load from the increased cell turnover driven by cured meats.
And their dosing is terrible as you point out, as a lot of the dose-response for the other endpoints are usually driven by each incremental 50 grams of consumption. Did they even calculate a dose response at all?
Good point. Definitely see studies linking CVD etc. to dementia.
I was startled not to find dose-response analyses in this one. They do have the data. This feels a bit like a researcher degrees of freedom problem, in the sense of their having chosen the analytic approach that yields the most favorable results.
Part of the reason I say that is the authors were noticeably vague about explaining the cut-offs for low, medium, and high intake. Their only justification is a reference to published standards "while also considering the distribution of intake among study participants." Ok, but what does that "also" mean? This fails one of those standards for scientific writing we teach undergrads, where you're supposed to describe your methods so clearly that someone else could replicate your work. Here too, I'm concerned that intake was simply trichotomized in a way that produced the strongest outcomes.
Can we PLEASE look at what they included in the study when they do this stuff. Did you notice they included "We used nationally-representative cross-sectional survey data to analyze consumption of unprocessed red meat; processed meat; and total red and processed meat."
In the study this includes "processed meat (defined as all meats that have been salted, cured, fermented, smoked, or otherwise processed for preservation and flavor enhancement)."
Flavor enchancement? "Additionally, flavor enhancers such as monosodium glutamate (MSG) and disodium 5'-ribonucleotides are used to amplify the savory taste of umami.
Hydrolyzed vegetable protein (HVP) is another flavor enhancer that brings out the savory, meaty flavor in processed meats, making it popular in vegetarian and vegan food products."
PROCESSED MEAT includes "Processed red meat refers to red meat that has been preserved through methods such as smoking, curing, salting, or adding chemical preservatives. This includes meats like ham, bacon, salami, and some sausages."
Adding chemical preservatives? Defined as "Processed meat often contains various chemical enhancements to improve its taste, texture, and shelf life. Common preservatives include sodium nitrates and sodium nitrites, which can increase the risk of heart disease and diabetes according to Mayo Clinic licensed dietician Kathrine Zeratsky.
Other additives include hydrolyzed protein, modified food starch, and monosodium glutamate (MSG).
These additives enhance flavor and texture but may introduce unnatural substances into the human body, potentially leading to harmful side effects and health issues.
Additionally, processed meats can contain artificial colorants, flavorings, and texturants, which are used to improve appearance and taste."
NOW tell me how accurate this study MAY BE????
Good point. Ambiguity as to what participants were actually eating is one reason the approach to measurement in this study was a hot mess. There's nothing in the study itself or the supplementary materials that adequately addresses your point. The food survey questions weren't (and probably couldn't be) specific enough.
I don't blame the researchers per se for the problem, because it's inherent to large observational studies - i.e., there's no way to accurately record what each of tens or hundreds of thousands of people eat each day over a period of years.
Practically speaking, it seems reasonable to avoid eating too much processed meat, on the grounds that something in it (though we can't say what) increases risk of health problems if we eat too much of it (though we can't be sure what the safe maximum would be).
The other thing is when people say “RED MEAT IS BAD FOR YOU” are they talking about the re meat sold at Wamart that is injected with “water and chemicals to increase its weight and improve its appearance. These chemicals can include salt, phosphates, antioxidants, and flavorings.”
OR
are they talking about your local grass-fed, grass finished farmer?
Usually it’s the WalMart crap!
Yeah... individual, large-n observational studies are rarely set up in a way that allows for such fine distinctions (e.g., processed red meats with high vs. low sodium content). Rather it's the totality of evidence (observational, experimental, petri dish, etc.) which suggests that red meat and salt, for instance, each increase health risks beyond certain intake levels.
Interesting until YOUR political biases ruined your credibility. You presented small bits of those political examples, much like the investigators didn’t report important data, or acknowledge it was missing. I’ll read the study myself.
I appreciate your point, but I think that my credibility should only be judged on the basis of how accurately and insightfully I reviewed the study, because: (a) I openly acknowledged my political biases, and (b) I presented evidence for my politically partisan complaints.
On occasion I air partisan complaints without evidence, but 100% of the time those are brief, snarky comments, and they have no bearing on my treatment of the data.
I wouldn't claim to be completely impartial - nobody is - but I do try my best to present nonpartisan discussions of study findings, and in some posts, like a recent one on butter and mortality, I end up criticizing elements of what might be called a liberal perspective.
As you point out so well, diet studies, especially those relying upon self-reporting, are fraught with errors. Meta-analyses suggesting trends, rather than single studies giving percentages, are far more persuasive, but don't seem to inspire as many headlines.
It's hard to get people interested in study methods. We're busy; we just want to know the takeaway.
I wish high school curricular standards made more room for supporting data literacy, because even kids at that age can appreciate the difference between one crappy observational study vs. a well-done meta-analysis (even if the stats are beyond what's covered in AP Statistics).
Countries such as Finland begin teaching critical thinking skills in elementary school. Just another area where the U.S. lags behind other developed nations.
Nice work and agree with your assessment of what seems to be an arbitrary and capricious criteria for the processed groups with rather low bars. You mentioned the "Informational siloing also seems to be occurring." This reminded me of the new proposed dietary guidelines that emphasise plant based over meat I shared earlier this month. Seeing the same thing, of course, thin s might have changed since then. Guess there is a global plot to take away your burger. Anxieties at the Petri Dish - by KB's FROM THE PETRI DISH
Next posting is going to look at a few things on diets, the gut biome and brain/ling/heart axis. Suspect there is an underlining commonality to this and other medical conditions.
Interesting plan for your next posting.
It's sad to see that issues like diet-health relationships are so deeply politicized. I understand that people are going to follow party lines when debating trade-offs between the economic benefits and the environmental damage created by the cattle industry, for instance, but it's a shame we can't just set politics aside when looking at the impact of beef consumption on health and let the data lead wherever it may.
Dementia or Alzheimers is an interesting endpoint, too, because cured meats like hot dogs and salami and pepperoni are fairly significant hazards for various cancers (1.1 - 1.4 HR), and of course red meat is also a significant CVD hazard (1.1 - 1.3 HR depending on unprocessed vs processed and consumption), likely due to saturated fat (and nitrites and nitrates in the processed / cured ones).
There's not necessarily any direct relations between those endpoints and dementia, but it definitely feels like there's probably additional dementia risk if you have CVD or higher cholesterol (and correspondingly higher inflammation), or if your body is under additional immunological and inflammatory load from the increased cell turnover driven by cured meats.
And their dosing is terrible as you point out, as a lot of the dose-response for the other endpoints are usually driven by each incremental 50 grams of consumption. Did they even calculate a dose response at all?
Good point. Definitely see studies linking CVD etc. to dementia.
I was startled not to find dose-response analyses in this one. They do have the data. This feels a bit like a researcher degrees of freedom problem, in the sense of their having chosen the analytic approach that yields the most favorable results.
Part of the reason I say that is the authors were noticeably vague about explaining the cut-offs for low, medium, and high intake. Their only justification is a reference to published standards "while also considering the distribution of intake among study participants." Ok, but what does that "also" mean? This fails one of those standards for scientific writing we teach undergrads, where you're supposed to describe your methods so clearly that someone else could replicate your work. Here too, I'm concerned that intake was simply trichotomized in a way that produced the strongest outcomes.
I was eating a cheeseburger whilst reading this. I'm reconsidering.
You'll be fine. Can't say the same for the cow. :)