What do you do when the latest clinical trial contradicts the core assumption of UPF science? How about accentuate the positive and eliminate the negative?
Viewing those images does nothing to reduce stereotypes about British food. I’d lose weight if you fed me that slop. Heck, the only way I’d eat that slop is if I were in prison.
Also, it all seems more processed than what I eat most of the time. Except for my daily protein shake laced with creatine.
You make a good point. Feed people food they don't like and they will eat less of it and maybe lose a little weight as these people did. Says nothing about the effect of the foods on the development of obesity (although why people don't like it might be worth learning). This kind of trial always reminds me of a friend I had 30 years ago who tried to quit smoking by getting acupuncture. When I asked her what the needle in her ear lobe was supposed to do, she said it would make her like smoking less. I said, "why don't you just buy cigarettes you don't like?" Needless to say, neither technique worked, although she ultimately quit smoking years before I managed.
Great article as always, Gary. As to your last line "...in the absence of the lengthy, expensive and perhaps impractical clinical trials that could provide definitive evidence," I as one loyal subscriber am, like you (I am guessing) gosh darn tired of the excuses from the bloated government agencies for why they don't actually do the necessary work. "Expensive?" YGBFKM! Ask the President for some of the impounded money saved from the dismantling of USAID, for crying out loud. Or divert some of the money saved from the departure of 25% of the IRS workforce. "Impractical?" I simply don't accept that assertion. It has never stopped agencies from conducting their confirmation-biased studies in the past. How about asking Mr. Kennedy, et al, - team MAHA - to be bold and brave here, and, to commit to a big worthy project, impracticalities and all?
It's well past time we stop settling for government agencies making rules based upon weak/nonexistent evidence. It's not too much to ask.
Yes, sadly we are in complete agreement here. Of course, then it gets complicated. Should they allocate the resources, then then have to decide which trials to do and which to do them. With a research community that has not only tolerated but embraced pseudoscientific concepts like UPFs, who do you trust to make those decisions? And who do you trust to carry out the trials? I have thought of ways to get around this problem and I even have a tiny bit of influence with the current administration--I know both Jay Bhattacharya and Marty Makary and both blurbed my last book on diabetes--but I can't imagine a political situation in which this problem is solved. I'll write about this more, I hope, in future posts.
I know I'm oversimplifying here, but whatever the formal, academic definitions of UPFs might currently be or come to be in the future, I can't help but think that when most laypeople hear the term "ultra-processed foods," what comes to mind are things like Pop-Tarts, Doritos, Frosted Flakes, and the like. A zero-sugar meat stick probably never even enters the picture.
Siobhan Huggins said something on stage at Low Carb Denver a few years ago (I think it was Denver anyway). It was sort of a casual comment, but it struck me as profound, and it's stayed with me since. I nominate this as a potential guiding principle for future research on UPFs: "I don't care how 'processed' something is; I care about the metabolic effects."
But then, I suppose, the question becomes, the metabolic effects in *whom?* (Hasn't previous research shown that people consuming the same quantity of the same foods can have vastly different responses with regard to glucose and insulin? Eran Segal's work comes to mind. Researchers "found high variability in the response to identical meals, suggesting that universal dietary recommendations may have limited utility." https://pubmed.ncbi.nlm.nih.gov/26590418/)
Another awesome piece Gary. You just keep churning them out! :-) There was lots to like here but, for me, the highlight was your section about the essence of bad science. I loved the Bing Crosby effect! These lessons apply much more broadly than nutrition science. They are especially relevant to the mental health and psychological distress field. It's mind-blowing to consider what sort of a world we might be living in if more scientists, more of the time, took a more impartial view of the evidence and were guided by the data rather than personal opinions and beliefs. I'm reminded of a quote I read in a Ben Goldacre book some years ago: "You can't reason people out of positions they didn't reason themselves into." :-)
Lost in the middle always leading to the "generally speaking" analysis. Nothing was always for everybody, regardless of the diet assigned. So we struggle to agree on the diet for everyone knowing full well that there will be those in the middle for which the agreed on diet will not apply for a variety of reasons - even when the participants are limited to healthy (whatever that is) people. Exactly like the results of any drug trial.
I don't even care who lost weight. I want to know who GAINED weight, the impact on the long list of health measurements (if any), and what their diet was (including an inventory of food in their pantry which was not supposed to be eaten), and as assessment of what perpetuated the gain.
I agree with you that no single diet is right for everyone clearly individuals can thrive on a wide range of different diets. But as I've argued in my books, I don't think the same is true for the cause of obesity itself, and I think if we get that science right, the diet question solves itself. So in a sense, knowing who gained weight in these studies would be interesting, but they still wouldn't tell us why.
Moreover, as I've also argued, you don't need government advice to find a diet that works for you. Most of us who struggle with our weight (I would think) try different dietary approaches hoping to find one that works. Most of us who ended up, so far, eating keto or even carnivore diets, did so because when we got to these approaches we found we could control our weight and mostly fix our chronic ailments. At this point, changing our diets because some authority says so seems silly, a return to a way of eating that was harmful... for us.
The problem with this conversation is definition of terms. Like, what exactly is an ultra processed food? Or is that even the best term. I think that any reasonable person knows that something is wrong with the modern American diet with obesity, diabetes, and hypertension being major symptoms. And, perhaps we can agree on some root causes which could stem from either ultra processed foods or minimally processed foods. Raw kale is an example of a minimally processed food that would actually be dangerous. And, I don't really think that raw vegan is such a good idea--talk about minimal processing. So, what is the root cause of the health crisis? Here are my candidates regardless UPF or MPF:
1. Foods/diets that cause inflammation.
2. Foods/diets that cause micronutrient deficiencies.
3. Foods/diets that cause caloric flooding (excessive calories that overwhelm the body's ability to process them safely.)
As an 82-year old who has been training for the Minnesota Senior Games--track and field (The event was this past August 9, and I "kicked butt," by the way.), I have been most interested in this third item--specifically how to properly fuel my body. Usual fuels are glucose and fatty acids. Glucose for (high intensity) anaerobic and fatty acids for (low to medium intensity) aerobic exercise--at medium intensity, I am told, they start becoming 50/50.
The problem I was having is that, while in training, during the final 100 meter sprint of the longer races, I would "hit the wall." For marathon runners this can happen at the 18-22-mile mark.
I am guessing that because I am a "fat burner" my glycogen storage ability might be smaller.
Then I read a study where they gave subjects 75 grams of glucose and measured blood glucose and lactate. Lactate showed up within 5 minutes and glucose not until 15 minutes.
So the small intestine is converting glucose to lactate, that is, I suppose, until that mechanism is overwhelmed. Further, I learned that lactate is actually the preferred source for energy production and can be used either in aerobic and anaerobic conditions. And, I am guessing that people who are carb dependent, do not burn up lactate and it simply accumulates, and later gets converted to glucose.
So, I started eating yogurt and drinking lightly fermented tart cherry juice before and during training. The "hitting the wall" went away.
Then during the track meet yesterday, running 50m, 800m, 100m 400m and 200m races within 2 plus hours, I even felt more energetic as time went on, and I finished the 200m with a personal best time. Recovery after each event went rather quickly
Now, because I couldn't measure what was going on metabolically this is all conjecture. But perhaps suggests some areas for investigation.
Hi Gary, Interesting thoughts and too much to get into here. I never quite get the focus on inflammation rather than endocrine response (insulin, primarily, glucagon, GH and now, obviously, incretins like GLP-1) and I think of these controlling fuel partitioning and why you would or would not hit the wall. I'm a fan of your solution, which is self-experimentation. Could you try, though, making only one change and seeing what happens: i.e., add yogurt only and see how that effects your performance? Than add the tart cherry juice only and see what happens? Might tell us whether it's both or only one and if only one, then knowing which one might tell us more about potential mechanisms.
I hate kale too, that's why I'm picking on it. The leaves of kale, like the leaves of rhubarb are very high in oxalates. "It binds minerals and has been linked to kidney stones and other health problems."
In the old days, kale was used as a base or background in the produce department. I think it is considered "healthy" because it is bulky.
To while away a hot Sunday afternoon, I asked chatgpt to summarize van Tulleken’s Ultra-Processed People (being too lazy to read the book myself). One of chatgpt's takeaways from the book was that that UPFs lead to overeating and obesity.
I then asked it to consider its answer in the light of the new study (always a good, and often amusing, idea to challenge chatgpt’s initial response). Here’s what came back…
************
That's a great question! It might seem contradictory at first glance, but there are a few ways to reconcile these findings. The key is to consider the context and specifics of each study. The general takeaway that ultra-processed foods (UPFs) are linked to obesity comes from observational studies showing associations across populations, often highlighting long-term dietary patterns and overall lifestyle factors.
In contrast, the recent study published in Nature Medicine was a controlled, randomized crossover trial where participants followed specific diets—either UPF or minimally processed—while adhering to healthy dietary guidelines.
************
If this is a fair summary of the book and the study, then to me it points – surprise, surprise – to the merits and difficulties of observational vs controlled dietary studies.
And thanks Gary, I now can't get that song out of my head...
Yeah, it's been in my head for a few days now, too. Luckily I like it a lot, so I don't mind (yet).
Re Chris van Tulleken, he had interviewed me for his book which includes a chapter about how charming and fit I am and also how wrong. I hadn't read it until this week when prompted by the Substack. So I'm wrong about sugar and carbs, according to Chris, because Kevin Hall has written papers arguing that I'm wrong.
I emailed Chris yesterday, included a link to this substack and asked him what I was missing. I also suggested we get on a Zoom and discuss that chapter and the wisdom of accusing someone of selective interpretive of the evidence (Bing Crosby science) while very clearly doing the same yourself.
If he gets back to me and we have any substantial discussion about it, I'll post an update.
I have some comments about your footnote 8, on the statistical analysis. You write, "As it turned out 12 of the original 55 participants dropped-out of the trial (every last one of them after being assigned to the MPF diet, which is head-scratching)." I don't quite understand this; subjects in this trial were assigned to a sequence of diets, UPF/MPF or MPF/UPF. To me, the interpretation depends on whether subjects mostly drop from the MPF/UPF sequence (during period 1) or they drop from the MPF diet in either time-point. However, in either case, the drop-outs result in missing data at end of trial. If the ITT analysis differed from the Per Protocol analysis, it must be a result of how the data were handled. I am surprised that the ITT analysis would be statistically significant when the Per Protocol analysis was not, because I would expect any modern analysis of this problem would end up in one of two places. Either the analysis would be qualitatively the same, except with larger confidence intervals and bigger p-values, or else it would be visibly worse than the Per Protocol analysis (geeky technical note: this is the Missing-Not-at-Random case; the fact that all the drop outs occurred on the MPF diet leads me to suspect that this is the applicable case.)
Here's a copy and paste from the Results section on page 2 of the study. "Six participants withdrew during the first-period MPF diet (sequence MPF/UPF), two during washout
(sequence MPF/UPF) and four during the second-period MPF diet (sequence UPF/MPF)."
Thank you for checking, Mr. Ellis. This actually raises a design question. Why is there a washout period? For a conventional cross-over design, you want time for drug treatments to wash out. But food? Since the subjects have to eat something during the washout, the idea that you can have a meaningful washout period is just silly.
Me again. I assume it means they return to their ‘usual’ diet. Hence the study is trying to test the change from usual to MPF (or usual to UPF) in the first period, and then the change from usual to UPF (or usual to MPF) in the second period. It assumes that four weeks is enough time to return to whatever constitutes normal for each participant.
I've tried to decipher the data presented in the supplementary tables. Tricky, because the data is unadjusted, and is presented as changes rather than absolute measurements, and so prone to me making errors (not unknown) trying to calculate the absolute data over the duration of the study - weight, waist-to height ratio, etc. But, with those caveats, it appears that there was a rebound in weight (for the first cohort) and waist-to-height ratio (for both cohorts) between the crossovers. Which if correct suggests to me that the unhealthiest part of the study was whatever the two groups were eating or doing in the four-week washout plus two-week re-baselining period.
It would also have been interesting if the researchers had included a third group, one asked to carry on "as normal" but to keep a food diary. This could have given a benchmark as to the presence of any research participation effect.
“it’s the insulin stupid” it just seems so obvious. The link of refined sugars direct effect on glycosylation of proteins in all organs and hyperinsulinemia adverse effects combine for a double whammy of multiple organ demise- especially in eyes, kidney, nerves, and vessels. it’s the carbs that turn our hormones against us. every time we mention refined carbs, sugar we need to mention insulin so awareness of high insulin is communicated. Every pt ( especially obese) should push their PCP. for an insulin level. a $20 test that is rarely ordered!!
I agree, Gary. I've always thought that the ultra-processed foods approach might have a moderating effect on the usage of carbs, even while agreeing that UPF seemed to be ill-defined and not nearly as useful as simply going ketogenic. Thank you for this article! -- Fritz
Thanks, Fritz. It might and the folks I know who are accepting of it think of it that way. Tell patients, for instance, not to eat UPFs or, even easier, not to eat packaged food of any kind, and they'll probably get a little healthier because they'll be avoiding some refined carbs and sugars (and maybe seed oils, dyes, etc.) But if the refined carbs and sugars are the problem, then aren't you better telling them to avoid those. That way they don't feel virtuous at the supermarket buying only from the produce and meat sections and go home and bake and eat minimally-processed cookies to celebrate.
Viewing those images does nothing to reduce stereotypes about British food. I’d lose weight if you fed me that slop. Heck, the only way I’d eat that slop is if I were in prison.
Also, it all seems more processed than what I eat most of the time. Except for my daily protein shake laced with creatine.
Hi Pete,
You make a good point. Feed people food they don't like and they will eat less of it and maybe lose a little weight as these people did. Says nothing about the effect of the foods on the development of obesity (although why people don't like it might be worth learning). This kind of trial always reminds me of a friend I had 30 years ago who tried to quit smoking by getting acupuncture. When I asked her what the needle in her ear lobe was supposed to do, she said it would make her like smoking less. I said, "why don't you just buy cigarettes you don't like?" Needless to say, neither technique worked, although she ultimately quit smoking years before I managed.
Great article as always, Gary. As to your last line "...in the absence of the lengthy, expensive and perhaps impractical clinical trials that could provide definitive evidence," I as one loyal subscriber am, like you (I am guessing) gosh darn tired of the excuses from the bloated government agencies for why they don't actually do the necessary work. "Expensive?" YGBFKM! Ask the President for some of the impounded money saved from the dismantling of USAID, for crying out loud. Or divert some of the money saved from the departure of 25% of the IRS workforce. "Impractical?" I simply don't accept that assertion. It has never stopped agencies from conducting their confirmation-biased studies in the past. How about asking Mr. Kennedy, et al, - team MAHA - to be bold and brave here, and, to commit to a big worthy project, impracticalities and all?
It's well past time we stop settling for government agencies making rules based upon weak/nonexistent evidence. It's not too much to ask.
Thanks again for the enlightening discourse.
Hi Jack,
Yes, sadly we are in complete agreement here. Of course, then it gets complicated. Should they allocate the resources, then then have to decide which trials to do and which to do them. With a research community that has not only tolerated but embraced pseudoscientific concepts like UPFs, who do you trust to make those decisions? And who do you trust to carry out the trials? I have thought of ways to get around this problem and I even have a tiny bit of influence with the current administration--I know both Jay Bhattacharya and Marty Makary and both blurbed my last book on diabetes--but I can't imagine a political situation in which this problem is solved. I'll write about this more, I hope, in future posts.
gt
I know I'm oversimplifying here, but whatever the formal, academic definitions of UPFs might currently be or come to be in the future, I can't help but think that when most laypeople hear the term "ultra-processed foods," what comes to mind are things like Pop-Tarts, Doritos, Frosted Flakes, and the like. A zero-sugar meat stick probably never even enters the picture.
Siobhan Huggins said something on stage at Low Carb Denver a few years ago (I think it was Denver anyway). It was sort of a casual comment, but it struck me as profound, and it's stayed with me since. I nominate this as a potential guiding principle for future research on UPFs: "I don't care how 'processed' something is; I care about the metabolic effects."
But then, I suppose, the question becomes, the metabolic effects in *whom?* (Hasn't previous research shown that people consuming the same quantity of the same foods can have vastly different responses with regard to glucose and insulin? Eran Segal's work comes to mind. Researchers "found high variability in the response to identical meals, suggesting that universal dietary recommendations may have limited utility." https://pubmed.ncbi.nlm.nih.gov/26590418/)
Another awesome piece Gary. You just keep churning them out! :-) There was lots to like here but, for me, the highlight was your section about the essence of bad science. I loved the Bing Crosby effect! These lessons apply much more broadly than nutrition science. They are especially relevant to the mental health and psychological distress field. It's mind-blowing to consider what sort of a world we might be living in if more scientists, more of the time, took a more impartial view of the evidence and were guided by the data rather than personal opinions and beliefs. I'm reminded of a quote I read in a Ben Goldacre book some years ago: "You can't reason people out of positions they didn't reason themselves into." :-)
Lost in the middle always leading to the "generally speaking" analysis. Nothing was always for everybody, regardless of the diet assigned. So we struggle to agree on the diet for everyone knowing full well that there will be those in the middle for which the agreed on diet will not apply for a variety of reasons - even when the participants are limited to healthy (whatever that is) people. Exactly like the results of any drug trial.
I don't even care who lost weight. I want to know who GAINED weight, the impact on the long list of health measurements (if any), and what their diet was (including an inventory of food in their pantry which was not supposed to be eaten), and as assessment of what perpetuated the gain.
Hi Pat,
I agree with you that no single diet is right for everyone clearly individuals can thrive on a wide range of different diets. But as I've argued in my books, I don't think the same is true for the cause of obesity itself, and I think if we get that science right, the diet question solves itself. So in a sense, knowing who gained weight in these studies would be interesting, but they still wouldn't tell us why.
Moreover, as I've also argued, you don't need government advice to find a diet that works for you. Most of us who struggle with our weight (I would think) try different dietary approaches hoping to find one that works. Most of us who ended up, so far, eating keto or even carnivore diets, did so because when we got to these approaches we found we could control our weight and mostly fix our chronic ailments. At this point, changing our diets because some authority says so seems silly, a return to a way of eating that was harmful... for us.
Best,
gt
The problem with this conversation is definition of terms. Like, what exactly is an ultra processed food? Or is that even the best term. I think that any reasonable person knows that something is wrong with the modern American diet with obesity, diabetes, and hypertension being major symptoms. And, perhaps we can agree on some root causes which could stem from either ultra processed foods or minimally processed foods. Raw kale is an example of a minimally processed food that would actually be dangerous. And, I don't really think that raw vegan is such a good idea--talk about minimal processing. So, what is the root cause of the health crisis? Here are my candidates regardless UPF or MPF:
1. Foods/diets that cause inflammation.
2. Foods/diets that cause micronutrient deficiencies.
3. Foods/diets that cause caloric flooding (excessive calories that overwhelm the body's ability to process them safely.)
As an 82-year old who has been training for the Minnesota Senior Games--track and field (The event was this past August 9, and I "kicked butt," by the way.), I have been most interested in this third item--specifically how to properly fuel my body. Usual fuels are glucose and fatty acids. Glucose for (high intensity) anaerobic and fatty acids for (low to medium intensity) aerobic exercise--at medium intensity, I am told, they start becoming 50/50.
The problem I was having is that, while in training, during the final 100 meter sprint of the longer races, I would "hit the wall." For marathon runners this can happen at the 18-22-mile mark.
I am guessing that because I am a "fat burner" my glycogen storage ability might be smaller.
Then I read a study where they gave subjects 75 grams of glucose and measured blood glucose and lactate. Lactate showed up within 5 minutes and glucose not until 15 minutes.
So the small intestine is converting glucose to lactate, that is, I suppose, until that mechanism is overwhelmed. Further, I learned that lactate is actually the preferred source for energy production and can be used either in aerobic and anaerobic conditions. And, I am guessing that people who are carb dependent, do not burn up lactate and it simply accumulates, and later gets converted to glucose.
So, I started eating yogurt and drinking lightly fermented tart cherry juice before and during training. The "hitting the wall" went away.
Then during the track meet yesterday, running 50m, 800m, 100m 400m and 200m races within 2 plus hours, I even felt more energetic as time went on, and I finished the 200m with a personal best time. Recovery after each event went rather quickly
Now, because I couldn't measure what was going on metabolically this is all conjecture. But perhaps suggests some areas for investigation.
Hi Gary, Interesting thoughts and too much to get into here. I never quite get the focus on inflammation rather than endocrine response (insulin, primarily, glucagon, GH and now, obviously, incretins like GLP-1) and I think of these controlling fuel partitioning and why you would or would not hit the wall. I'm a fan of your solution, which is self-experimentation. Could you try, though, making only one change and seeing what happens: i.e., add yogurt only and see how that effects your performance? Than add the tart cherry juice only and see what happens? Might tell us whether it's both or only one and if only one, then knowing which one might tell us more about potential mechanisms.
Why is raw kale dangerous?
That’s a purely intellectual question; I hate kale.
I hate kale too, that's why I'm picking on it. The leaves of kale, like the leaves of rhubarb are very high in oxalates. "It binds minerals and has been linked to kidney stones and other health problems."
In the old days, kale was used as a base or background in the produce department. I think it is considered "healthy" because it is bulky.
To while away a hot Sunday afternoon, I asked chatgpt to summarize van Tulleken’s Ultra-Processed People (being too lazy to read the book myself). One of chatgpt's takeaways from the book was that that UPFs lead to overeating and obesity.
I then asked it to consider its answer in the light of the new study (always a good, and often amusing, idea to challenge chatgpt’s initial response). Here’s what came back…
************
That's a great question! It might seem contradictory at first glance, but there are a few ways to reconcile these findings. The key is to consider the context and specifics of each study. The general takeaway that ultra-processed foods (UPFs) are linked to obesity comes from observational studies showing associations across populations, often highlighting long-term dietary patterns and overall lifestyle factors.
In contrast, the recent study published in Nature Medicine was a controlled, randomized crossover trial where participants followed specific diets—either UPF or minimally processed—while adhering to healthy dietary guidelines.
************
If this is a fair summary of the book and the study, then to me it points – surprise, surprise – to the merits and difficulties of observational vs controlled dietary studies.
And thanks Gary, I now can't get that song out of my head...
Hi David,
Yeah, it's been in my head for a few days now, too. Luckily I like it a lot, so I don't mind (yet).
Re Chris van Tulleken, he had interviewed me for his book which includes a chapter about how charming and fit I am and also how wrong. I hadn't read it until this week when prompted by the Substack. So I'm wrong about sugar and carbs, according to Chris, because Kevin Hall has written papers arguing that I'm wrong.
I emailed Chris yesterday, included a link to this substack and asked him what I was missing. I also suggested we get on a Zoom and discuss that chapter and the wisdom of accusing someone of selective interpretive of the evidence (Bing Crosby science) while very clearly doing the same yourself.
If he gets back to me and we have any substantial discussion about it, I'll post an update.
I have some comments about your footnote 8, on the statistical analysis. You write, "As it turned out 12 of the original 55 participants dropped-out of the trial (every last one of them after being assigned to the MPF diet, which is head-scratching)." I don't quite understand this; subjects in this trial were assigned to a sequence of diets, UPF/MPF or MPF/UPF. To me, the interpretation depends on whether subjects mostly drop from the MPF/UPF sequence (during period 1) or they drop from the MPF diet in either time-point. However, in either case, the drop-outs result in missing data at end of trial. If the ITT analysis differed from the Per Protocol analysis, it must be a result of how the data were handled. I am surprised that the ITT analysis would be statistically significant when the Per Protocol analysis was not, because I would expect any modern analysis of this problem would end up in one of two places. Either the analysis would be qualitatively the same, except with larger confidence intervals and bigger p-values, or else it would be visibly worse than the Per Protocol analysis (geeky technical note: this is the Missing-Not-at-Random case; the fact that all the drop outs occurred on the MPF diet leads me to suspect that this is the applicable case.)
Here's a copy and paste from the Results section on page 2 of the study. "Six participants withdrew during the first-period MPF diet (sequence MPF/UPF), two during washout
(sequence MPF/UPF) and four during the second-period MPF diet (sequence UPF/MPF)."
Thank you for checking, Mr. Ellis. This actually raises a design question. Why is there a washout period? For a conventional cross-over design, you want time for drug treatments to wash out. But food? Since the subjects have to eat something during the washout, the idea that you can have a meaningful washout period is just silly.
Me again. I assume it means they return to their ‘usual’ diet. Hence the study is trying to test the change from usual to MPF (or usual to UPF) in the first period, and then the change from usual to UPF (or usual to MPF) in the second period. It assumes that four weeks is enough time to return to whatever constitutes normal for each participant.
Thanks, David for the copy and paste.
As for the wash-out rationale, I described it in this post in the context of the "landmark" NIH trial that did not include one: uncertaintyprinciples.substack.com/p/nih-has-a-nutrition-problem-part
Yes, I agree this is surprising.
I've tried to decipher the data presented in the supplementary tables. Tricky, because the data is unadjusted, and is presented as changes rather than absolute measurements, and so prone to me making errors (not unknown) trying to calculate the absolute data over the duration of the study - weight, waist-to height ratio, etc. But, with those caveats, it appears that there was a rebound in weight (for the first cohort) and waist-to-height ratio (for both cohorts) between the crossovers. Which if correct suggests to me that the unhealthiest part of the study was whatever the two groups were eating or doing in the four-week washout plus two-week re-baselining period.
It would also have been interesting if the researchers had included a third group, one asked to carry on "as normal" but to keep a food diary. This could have given a benchmark as to the presence of any research participation effect.
Ignore my comment on waist-to-height for now - I've decided I don't understand the height and waist measurements in the supplementary data.
What would be really useful from the researchers are the actual weight, BMI and waist measurements at every stage of the study. Or have I missed them?
“it’s the insulin stupid” it just seems so obvious. The link of refined sugars direct effect on glycosylation of proteins in all organs and hyperinsulinemia adverse effects combine for a double whammy of multiple organ demise- especially in eyes, kidney, nerves, and vessels. it’s the carbs that turn our hormones against us. every time we mention refined carbs, sugar we need to mention insulin so awareness of high insulin is communicated. Every pt ( especially obese) should push their PCP. for an insulin level. a $20 test that is rarely ordered!!
Yeah, moral license is a powerful effect!
I agree, Gary. I've always thought that the ultra-processed foods approach might have a moderating effect on the usage of carbs, even while agreeing that UPF seemed to be ill-defined and not nearly as useful as simply going ketogenic. Thank you for this article! -- Fritz
Thanks, Fritz. It might and the folks I know who are accepting of it think of it that way. Tell patients, for instance, not to eat UPFs or, even easier, not to eat packaged food of any kind, and they'll probably get a little healthier because they'll be avoiding some refined carbs and sugars (and maybe seed oils, dyes, etc.) But if the refined carbs and sugars are the problem, then aren't you better telling them to avoid those. That way they don't feel virtuous at the supermarket buying only from the produce and meat sections and go home and bake and eat minimally-processed cookies to celebrate.