[hifromwork]

It looks like they asked (a large number of) participants what they eat, and checked a correlation between their health and how closely they follow their Planetary Health Diet[1].

First, I don't see any attempt to control for the fact that people who eat "healthy" diets (vegetarian, vegan, avoid sugars and red meat, etc) are often the same people who care about exercise, do routine health checks etc. So even if the diet itself was a placebo, this mere correlation would show results as if it's helpful.

Second, It's already a known fact that avoiding too much red meat and sugar is good for you, so it's not surprising that a mostly vegetarian/vegan diet has health benefits compared to the average. I guess this study shows that this diet, which was designed for planet health, is also good for human health? It's a noble goal, but most people - maybe selfishly - primarily care about their own health. It would be interesting to quantify how much worse for health this diet is compared to a diet optimised for human health[2].

I also don't get the 30% lower CO2 part. Is it about CO2 generated by growing food/raising livestock? One doesn't need a study to know that eating mostly plants generates less CO2.

Disclaimer: I don't have access to the full paper so I've only studied the available abstract.

[1] Mostly vegan with some animal protein: https://eatforum.org/eat-lancet-commission/the-planetary-hea...

[2] Ideally it is (within an experimental error) as good as the "optimal" human diet, but it would be nice to hear that explicitly.

[xenonite]

About your third point: raising animals on pastures is not necessarily worse regards climate compared to growing vegetables on fields. It may well be the other way around. For example: the humus layer is much deeper with grass, hence storing more CO2.

About your second point, you name it as "fact" that too much red meat is problematic. Scientists are not so sure if it is about the meat or about some side-effect, like e.g., a virus transmitted along with (rare/raw, or even higher heated) meat in Western societies. Reference: https://doi.org/10.1002/ijc.27413

[piva00]

> About your third point: raising animals on pastures is not necessarily worse regards climate compared to growing vegetables on fields.

Thermodynamics by itself would already question this assertion, no?

[rickydroll]

maybe. there is an emerging body of evidence showing that different grazing methods influence carbon sequestration rates. https://pastureproject.org/about-us/regenerative-grazing-ben... https://www.theguardian.com/environment/article/2024/may/15/...

[hifromwork]

>About your second point, you name it as "fact" that too much red meat is problematic.

Too much red meat is tautologically problematic, because too much of anything (including water) is problematic :). I'm not a dietician so I don't have opinion on this, but I was thinking mostly of a common western diet full of highly-processed, highly fat and overly salty meats.

[nradov]

It is not at all a "known fact" that "avoiding too much red meat" is "good for you". There has never been a single large-scale, long-term randomized controlled trial which has shown such causality.

In general there are hardly any "known facts" in biology, least of all in human nutrition. All we have are probabilities. Most human nutrition studies that relate to this issue have been observational and relied on subject-reported data: in other words, junk science.

If there is an actual signal here one way or the other then it certainly has a much smaller impact on overall human health than other factors like energy balance, exercise, sleep quality, chronic stress, etc. There's a lot of other stuff to optimize first before we even think about the relative quantities of plant/fungus versus animal foods in our diets.

https://peterattiamd.com/qps1/

[adrian_b]

What I would have liked to see and which should exist in the full paper, but it is not described in the summary or in the snippets, is the exact method used to compute the score for a given diet.

It is not possible to compute any kind of valid score for an aliment taken in isolation, because no aliment is healthy if one would eat only that.

The various aliments complement their contents of nutrients and a pair of certain aliments can be very healthy, even when eating only one in the absence of the other would be unhealthy.

Moreover the quantity of an aliment matters a lot. For small quantities, most aliments are neither healthy nor unhealthy, they do not provide any noticeable contribution, positive or negative. For medium quantities, an aliment can be healthy or unhealthy, depending on which are the other components of the diet. For large quantities, almost any aliment becomes unhealthy.

So computing a realistic score for a daily diet is quite difficult and just giving vague guidelines like eating "mostly" vegetables with "a little" animal protein, while correct is completely unhelpful for providing any quantitative conclusions.

I hope that the full paper includes a valid methodology for characterizing a diet, but I somewhat doubt this, because I have never seen such a precise methodology yet.

The first step for computing a score would be to have for each component of the daily diet both the daily intake and the array of values with its content for all of the about 50 essential nutrients that are required by a human to live. The content values should take into consideration the digestibility of that aliment by humans.

Then by multiplying the nutrient contents with the daily intakes, one would obtain an array of about 50 values with the daily intakes for each essential nutrient. For each nutrient there is an optimal range for the daily intake, and the values that are either higher or lower must be penalized the farther they are from the optimal range.

To this initial score, various correction factors must be applied. The food must require a certain amount of chewing effort, in order to preserve the health of the teeth. If that is not true a diet must be penalized. The food must contain some amount of indigestible fiber, to help the transit through the intestine. If that is not true a diet must be penalized. There are many vegetables or fruits for which there is decent evidence that they contain something that improves health, especially cardiovascular health, but it is unknown which are the exact substances with favorable effect and which is the mechanism that explains their action. A diet containing such vegetables or fruits should be scored better.