All that’s plant-based is not meat

A new study highlights some of the pitfalls in plant-based diets, as well as some of the pitfalls in science journalism.

Plant-based products have grown in popularity as new, “meat-like” alternatives. Many companies and restaurant chains have jumped onto this so-called plant-based “meat” alternative (PBMA/PBMD for “diet”) trend to provide a wider variety of meats while also supplanting typical meat-based diets.

This has led to interest in PBMA as an allegedly sustainable method of eating as argued by environmental/climate change activists, but this has also led to criticisms over the actual nutritive, beneficial, and sustainable nature of these products.

This is mainly owed to the fact that many of these products require multiple layers of processing, as well as the addition of fillers/additives to obtain the proper texture and flavor that mimics meat. PBM also require fortification due to the otherwise low nutritional content, which begs the question if fortified PBMA is any better than nutrients sourced from real, whole foods.

This has been my general perspective on PBM- at the point that so much chemistry and science is required to obtain a meat-like product why not just eat the meat instead? And as more people begin to question the degree of ultra-processed foods in our standard diet is there even room on the table for processed albeit plant-based products?

It’s important to note that PBMA should be considered separate from a plant-based diet which may comprise whole, highly nutritive produce and is likely to be far superior to PBMA.

So, from this perspective we may not look too kindly at PBMA.

And this may be substantiated by a new study¹ which examined differences in cardiometabolic measures between people who ate animal-based meat and those who ate PBMA.

Such a study is pertinent in providing answers regarding the actual benefits of PBMA, and on the surface the study doesn’t appear to suggest a gross benefit in swapping out meat with plants. However, let’s explore some of the methodology and see if there are some problems. Remember that just because we agree with the conclusion of a study doesn’t mean that the researchers were able to reach that conclusion without issues.

---

Singapore PBMA Study

*Click on the above heading for another link to the study.

The study was conducted in Singapore using a small number of participants (~40 for both the meat and plant groups). Participants were randomized into either the meat group or the PBMA group and asked to eat the respective products over the course of 8 weeks. This included breakfast, lunch, dinner, and a snack.

There was unfortunately not much variety in the provided meat/plants. The authors categorize the products into 6 broad categories:

• Beef mince (ground beef)

• Pork mince (ground pork)

• Chicken breast

• Burger patty

• Sausage

• Chicken nuggets

What a sham! Not even a steak in sight! It’s also weird that a burger patty is categorized separate from ground beef. But it may not be too big of a loss; participants were provided locally sourced meats aside from chicken nuggets which were commercially sourced.

For the people within the PBMA group they were provided commercially available alternatives:

Corresponding to this list, the PBMD group was provided with the following foods: (1) Impossible Beef (Impossible Foods), (2) OmniMeat Mince (OmniFoods), (3) Chickened Out Chunks (The Vegetarian Butcher), (4) Beyond Burger (Beyond Meat), (5) Beyond Sausage Original Brat (Beyond Meat) and (6) Little Peckers (The Vegetarian Butcher).

Sounds appetizing, I suppose... Participants were provided these foods frozen in 3-day cycles, with protein content being similarly matched between both the animal and PBMA group. Participants were also asked to keep the rest of their diet relatively the same although participants were asked to reduce their protein intake from other sources outside of the provided protein:

Throughout the 8-weeks, participants were encouraged to minimize their consumption of other protein-rich foods (≤ 1 serving per 3-day cycle) beyond the intervention foods provided. The mode of preparation for intervention foods including the method of cooking, type of seasoning used, and meal accompaniments were at the discretion of the participants although as much as possible, participants were instructed keep the other components of their habitual diet consistent (e.g., staple foods, fruits, vegetables). Hedonic acceptability of the foods provided (in terms of appearance, taste, aroma and texture) and ease of dietary incorporation were evaluated using a continuous visual analogue scale after the 8-week dietary intervention.

Unfortunately, this poses lots of problems for the study as it introduces a ton of variability by way of individuals likely having different dietary behaviors. That is, the type of fruits, vegetables, herbs and seasonings that one person uses will likely differ greatly from someone else. Also, not all people consume the same number of meals (participants were asked to eat 3 meals a day as well as a snack), and people may consume different levels of protein and sources as well.

Note that participants were included if they consumed between 2 and 4 servings of protein a day (around 20 grams protein/serving) and so even the original protein consumption of the participants may have differed from what is expected from the diet regimen proposed by the researchers.

All to say that not much controlling was conducted with this study, and therefore other variables should be considered with respect to the measures taken. Now, the researchers do note that participants were asked to fill out occasional 3-day self-reports on their dietary habits and were coached on how to comply with their corresponding dietary group, however this information is left ambiguous (they appear to only suggest how to comply with eating their corresponding protein source).

Issues also extend into dietary measures from before the study (Week 0) and after the study (Week 8). Measures of carb, protein, and fat intake were based on participants’ dietary self-reports and were corresponded with nutritional databases, and thus are likely skewed based on these references. Nonetheless, note that baseline measures are not similar across the two groups, with the PBMA group appearing to have higher baseline dietary fat, carb, and calorie intake relative to the baseline measures of the animal group:

From Kiat Toh, et al. Boxed above are noticeable differences in baseline characteristics between the animal (ABMD) and plant (PBMD) protein groups such that those within the PBMD group had higher calorie, higher fat, and higher carb/sugar intake relative to the ABMD group.

And this is also reflected in some of the cardiometabolic biomarkers measured to the extent that those within the PBMA group appeared to have higher levels of cholesterol, C-reactive protein, and had a higher Framinghan 10-year CVD risk score at baseline:

These measures may not be too different between the groups, but again it reinforces the point that the health and dietary behaviors among individuals within each group may be different. It’s possible that these may show some sort of correlation between a possible unhealthier eating habit of those within the PBMA to begin with, and therefore minimal changes between the groups should be balanced by the fact that PBMA products may not be that much better than real meat, or that those within the PBMA are already starting at a point with worse health markers in general relative to those within the animal group.

There is one additional point of contention, although this seems to be a more controlled measure relative to the rest of the study. Along with the 8-week dietary restriction participants were also asked to participate in chronic glucose monitoring for 14 days. The first 3 days comprised a full-feeding period in which participants ate one of the provided frozen products along with a corresponding carbohydrate:

As a part of the 14-day CGMS period, participants first completed a full-feeding period, that spanned from day 0 dinner to day 3 dinner. This comprised of 13 meals including breakfast (0800 h), lunch (1200 h), snack (1600 h) and dinner (2000 h) that were consumed at fixed timings daily. Apart from the snack meal, participants cooked and consumed 1 of the 6 frozen ‘meats’ provided, with a fixed staple that included a serving of either white rice (210 g; HeatBahn, CJ Foods), hamburger bun (55 g; Gardenia hamburger buns, Gardenia Foods Pte Ltd) or plain instant noodles (70 g; Koka non-fried plain instant noodles, Tat Hui Foods Pte Ltd). The type of frozen ‘meat’ consumed between groups was congruent and protein-matched, with an identical snack eaten on all three days.

The language here doesn’t suggest that participants were allowed to eat other foods (the specifics of the diet are in the Supplemental material which is unfortunately unavailable), but keep in mind this possibility.

What’s interesting is that chronic glucose monitoring appears to suggest that participants within the PBMA group had higher glucose peaks and fell above the acceptable range for longer relative to the animal group. This is reflected in Figure 2 which shows the average glucose levels between the two groups after the first day of the full-feeding period:

Interstitial glucose levels among participants within both the PBMD and animal meat group. The measures reflect the first day/24-hour period of a full-feeding period. Note that interstitial glucose levels reflect levels of glucose within the interstitial space right below the skin and therefore may not reflect actual blood glucose levels.

Now, one possible explanation for these spikes is the fact that PBMA products tend to contain higher carbohydrate levels, and thus may reflect the overall higher consumption of carbohydrates among participants within this group. But once again, baseline measures for the PBMA group already reflect the fact that the PBMA group appears to consume higher levels of carbohydrates and sugar anyways, and this may be carryover from that fact. Note that over the 3-day full-feeding period the AUC for the PBMA group continuously declined while the AUC for the animal group appeared consistent (well, aside from the increase on Day 3 for some reason):

The AUC for a full-feeding period in which participants had continuous glucose monitoring. While the animal-based group had relatively consistent AUC measures across the 3 days the PBMD group showed a higher starting AUC that lowered over time, possibly reflecting the relatively higher carbohydrate/sugar intake of the PBMD group. However, this isn’t reflected within the integrated AUC measures which are more sporadic. I’m not sure the difference between the two measures, and bear in mind that the overall carb intake may not be the same across the 3 days as some of the plant-based products may be higher in carbs than others.

So maybe the higher peaks are an artifact of the generally higher carbohydrate consumption of the PBMA group? Not sure, but it’s at least something to consider.

Key Takeaways

I may have mentioned this previously, but I have a friend who had a moment of being vegetarian (he was dating a vegan so maybe that was the problem…). I recall one time where he and the girl talked about all the vegetarian foods that they ate. Vegetarian mozzarella sticks, vegetarian fish sticks, etc. etc. Even back then in my naïve state the list mentioned sounded weird. I mean, how exactly would you make nonfish taste like fish? It’s probably an example of how just because you can doesn’t mean you should. I’m also reminded of Ellen DeGeneres going through her vegan phase where she brought on someone to talk about the greatness of soy crumbles- just sprinkle soy crumbles into anything that would use ground beef and it’s allegedly the same thing!

We’re at this strange point where this push for less animal products comes with a push for highly processed, expensive foods that only serve to mimic the real deal. It’s the irony that less is more in this case means less meat and more additives and fillers.

In my opinion this study doesn’t appear to provide a clear indication that PBMA may be worse than animal products, but it also doesn’t appear to suggest it’s any better. There is also a slew of confounders which make extrapolation of this data difficult especially given that participants continued to eat their typical diet but swapped out animal products with plant-based products. And even then, the animal products provided may not be representative of products omnivores may typically eat (seriously, where’s the steak?!), and the degree of PBMA products provided may not represent the typical degree of PBMA that people may consume overall, even if it may represent a meatless society’s approach to eating.

The glucose peaks and GRADE values also must consider the fact that the PBMA group had higher baseline carbohydrate and sugar intake relative to the animal group, and even by the 8-week mark still had relatively higher carbohydrate intake. This would contradict the higher carbohydrate content of PBMA since the overall decrease in carbohydrate intake among the PBMA group during the 8-week study would suggest that these participants may have been sourcing protein from foods that were also relatively high in carbohydrates. Reliable recall, accurate accounting of nutrition, and biases should be considered that may bring into question some of the dietary measures provided.

A far better study would have been to follow the restricted diet proposed for the first full-feeding period/glucose monitoring portion of the study. Restrict the diet so that only the protein source will vary between the two groups. Granted, such a restricted study is likely not to be well tolerated- the ability to eat other foods may just be a respite from having to eat the plant-based meat products… But it would nonetheless provide for stricter variable control and better assessment of data.

If anything, studies such as these raise the importance of understanding what exactly is in our food. Note that PBMA products tend to contain much higher amounts of sodium and potassium, likely reflecting the fact that these products still must obtain a relatively stable shelf life like other frozen, ultra-processed foods. Thus, it probably wouldn’t be favorable to eat these sorts of foods if you are trying to avoid processed foods in general or anything with high sodium.

Again, we must emphasize the fact that PBMA are not similar to whole foods. Whole plants and less processed animal products are likely to be far more better and contain less additives than plant-based alternatives.

Compare this table from a study cited by Kiat Toh, et al. which compares the content of animal products to those of PBMA used in their study (Crimarco, et al.²):

From Crimarco, et al. A comparison of ingredients in animal and plant-based products used by the researchers. These products may not be the same as the ones used in the Kiat Toh, et al. study but it reflects the fact that plant-based products may be full of additional ingredients such as additives and fillers.

Even if all these additives are relatively safe to eat (most are things you can find in other foods generally) someone may not want all of these additives in one product, especially if it’s intended to make the product taste like beef or chicken.

And if you are trying to avoid carbohydrates in general then PBMA products likely won’t work as viable substitutes.

Thus, so far it appears that this push for plant-based meat may just be a wash in the long run.

Where’s the Study?

Given some issues with this study it’d be farfetched to extrapolate a ton of information from this study. And yet, I was rather surprised to see such videos as this one from Fox News:

As stated before this study may just suggest that PBMA isn’t any more healthful compared to real meat, but does it really suggest a negative impact on heart health from eating PBMA? Not quite given some of the results.

What’s even more surprising is that in the video above there seems to be something missing. As in, there doesn’t appear to be any discussion about the actual study. The only thing that is shown is the following image, with further discussion detailing the benefits of real meat compared to PBMA:

Sure, I certainly think that real meat is far more beneficial than a simulacrum, but again did this study show higher blood sugars when the PBMA had higher consumption of carbs and sugar even before the study? And where did the higher risk for heart disease come from? Nowhere in the study did the researchers mention heart disease or diabetes aside from suggesting that the included participants were those who were at risk of developing diabetes.

And this is my main problem with this news report. Instead of covering the actual study it takes what appears to be a misinterpretation and extrapolates talking points. Even if the talking points may be correct it still doesn’t mean that the inferences made regarding this study are sound.

And this doesn’t appear to be the only place to have misinterpreted the study.

Note this article published in The Western Journal

• which links to a Daily Mail article

• which then cites this article from News Medical

I think it’d be interesting for readers to see where some of the holes in the reporting lie, but what’s fascinating (or maybe shocking) is to see how readily outlets may not elaborate or breakdown a study, but instead may rely on secondhand accounts of other people’s reporting as a way of assessing a study.

This creates serious problems when one may misinterpret the findings of a study, leading to propagation and more misinterpretations down the line.

But this more of a problem not just in mainstream science coverage but of coverage in alternative outlets as well where journalists may be more inclined to look anywhere but the study and primary source.

So right now there appears to be an unfortunate reactionary effect where people may be taking this study to argue the dangers of plant-based meat when it just suggests that there may be no overall benefit.

It’s a reminder that not all studies are good studies, and we should be careful of praising a study if it falls in line with our narrative or position.

---

If you enjoyed this post and other works please consider supporting me through a paid Substack subscription or through my Ko-fi. Any bit helps, and it encourages independent creators and journalists such as myself to provide work outside of the mainstream narrative.

Click on the image above or visit the website at https://ko-fi.com/moderndiscontent

Photo by amirali mirhashemian on Unsplash

1. Kiat Toh, D. W., Fu, A. S., Mehta, K. A., Lin Lam, N. Y., Haldar, S., & Henry, C. J. (2024). Plant-based meat analogues (PBMAs) and their effects on cardiometabolic health: An 8-week randomized controlled trial comparing PBMAs with their corresponding animal-based foods. The American journal of clinical nutrition, S0002-9165(24)00396-4. Advance online publication. https://doi.org/10.1016/j.ajcnut.2024.04.006

2. Crimarco, A., Springfield, S., Petlura, C., Streaty, T., Cunanan, K., Lee, J., Fielding-Singh, P., Carter, M. M., Topf, M. A., Wastyk, H. C., Sonnenburg, E. D., Sonnenburg, J. L., & Gardner, C. D. (2020). A randomized crossover trial on the effect of plant-based compared with animal-based meat on trimethylamine-N-oxide and cardiovascular disease risk factors in generally healthy adults: Study With Appetizing Plantfood-Meat Eating Alternative Trial (SWAP-MEAT). The American journal of clinical nutrition, 112(5), 1188–1199. https://doi.org/10.1093/ajcn/nqaa203