Kevin Hall: Common Ground on Ultra-Processed Foods?

The opening of the Sixth Canadian Obesity Summit was a bit of a homecoming for Kevin Hall. It was his first visit to Ottawa in about three decades and Obesity Canada rewarded him with its Distinguished Lecturer Award. He used the occasion to review the evidence behind the diet wars wars that rage on Twitter. And he had a sensible proposition to offer. Instead of fighting about carbs and fat, can’t we find common ground on the problems that ultra-processed foods seem to present?

Restricting Carbs?

Right now, pop nutrition is telling us that ketogenic diets offer the hottest ticket for weight control. And in fact, David Ludwig proposes that carbohydrates and insulin are key drivers for our present problems with obesity. His thinking and his publications hold that carbohydrates – especially highly refined carbs – have become so common in Western diets that they are driving insulin secretion to unhealthy levels. In turn, this produces weight gain.

As he’s done before, Hall offered substantial doubts about Ludwig’s model. Those doubts are not especially welcome among folks with a big personal investment in ketogenic diets and carbohydrate restriction.

Reflecting on his data and a systematic review of the literature, Hall concluded that neither restricting fats nor carbs offers a compelling answer:

What this tells me is that the body does a pretty darn good job of adapting. It is more or less as efficient on these different diets. It’s amazing. Evolution made this to be so. We can survive in a variety of different environments, eat a variety of different diets, widely varying in carbohydrates and fats. It’s not magic.

Diet Quality and Calories?

But his new research, now accepted for publication in Cell Metabolism, suggests a third way – focusing on the quality of unprocessed and ultra-processed foods. Ultra-processed foods seem to prompt people to consume more calories. Hall summarized:

Is it the carbs, is it the calories, or is it the quality? This last study seems to suggest that the quality of the foods influences weight via the calories that people eat.

Controlling calories and manipulating carbs versus fat don’t seem to influence body fat change.

What distinguishes ultra-processed foods? People eating highly processed foods got more refined carbs and more fat. In the end, those foods seemed to prompt people to consume more calories.

How different are these two views? Refined carbs seem to be a problem from both perspectives. While the low-carb warriors insist that it’s all about the carbs and insulin, Hall is  focusing on quality and calories.

Do we have the data necessary to close the gap in these two viewpoints? Or is the real problem one of intellectual bias?

Click here for Hall’s slides and here for a preprint of his paper to be published in Cell Metabolism. For video of the opening ceremonies, including Hall’s lecture (@24 min), click here.

Trix, photograph © Mike Mozart / flickr

Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.


April 24, 2019

5 Responses to “Kevin Hall: Common Ground on Ultra-Processed Foods?”

  1. April 24, 2019 at 7:05 am, Anthony Pearson, MD, FACC said:

    Thank you for summarizing the latest thinking from Dr. Hall, who is doing really important research in this area. Greatly appreciate the link to his slides and the preprint.

  2. April 24, 2019 at 8:36 am, Mary-Jo said:

    I agree with the comment above. Measuring energy intake as an outcome of ultra-processed vs unprocessed diet is an overdue important key to insight of relationship of diet to effects to the body, possibly obesity. I found it interesting that subjective factors, such as fullness, satisfaction, etc. showed no significant differences between the two types of diet. As a dietitian, I was always touting the ‘greater satiety’ one would get from eating more wholesome foods and preparing meals from fresh ingredients vs. ready-to-eat whilst often acknowledging to patients that, yes, in the end you will most likely consume less calories, additives, fillers, and ‘stuff’ that’s not needed by the body, even though, of course, this was just intuitive thinking, on my part. Now, there’s some nice evidence affirming this is likely happening. Thanks for the summary and links and looking forward to more to come.

  3. April 24, 2019 at 10:15 am, David Brown said:

    “Do we have the data necessary to close the gap in these two viewpoints?” Yes. “Or is the real problem one of intellectual bias?” No, the problem is ignorance. Neither side of the debate is paying attention to important aspects of lipid research. What’s being ignored is the changes in the fatty acid profile of the food supply as set forth in this paragraph: We now know that major changes have taken place in the food supply over the last 100 years, when food technology and modern agriculture led to enormous production of vegetable oils high in ω-6 fatty acids, and changed animal feeds from grass to grains, thus increasing the amount of ω-6 fatty acids at the level of LA (from oils) and arachidonic acid (AA) (from meat, eggs, dairy).[1]

    In a 2003 Orlando Sentinel article entitled ‘Hunger Confronts Bigger Issue’, global obesity expert Barry Popkin repeatedly mentioned the global increase in vegetable oil (for food preparation) and meat consumption:

    “If you go back to those same villages or slum areas today … their diet includes a lot of vegetable oil … In China … Rice and flour intake is down, and animal-source foods such as pork and poultry and fish are way up, and the steepest increase is in the use of edible vegetable oils for cooking … People are eating more diverse and tasty meals; in fact, edible oil is a most-important ingredient in enhancing the texture and taste of dishes … The edible-oil increase is found throughout Asia and Africa and the Middle East as a major source of change.”[2]

    South African researcher Annadie Krygsman notes that “This risk of developing obesity and T2D has largely been blamed on the increased consumption of energy dense foods and fat intake, particularly saturated fat, but it is interesting to know that the mean fat intake of the human population has not increased much in the past 50 years.It is true that the vast advancement in technological developments has led to a reduction in physical activity worldwide, but as obesity now involves infants and the populations of developing countries, this obesity pandemic cannot be attributed to this alone.In addition, laboratory and other domesticated animals have also been subject to the increased prevalence of obesity, despite having largely unchanged living conditions for many years… The main alteration within the fatty acid profile of the modern diet has been the increased use of vegetable oil, both as a cheaper and more accessible alternative to animal fats, but also as a substitute to animal fats to reduce the intake of SFAs. Vegetable oils, although higher in monounsaturated fatty acids, are very high in omega-6 fatty acids (100-fold larger) compared with animal fat.These oils are also much lower in omega-3 fatty acids, leading to an increase in the ratio of dietary omega-6:omega-3 fatty acids. In the past few years a growing number of studies and meta-analyses have focused on the influence of dietary omega-6:omega-3 ratio, and the role of omega-3 fatty acids in the development of T2D and CHD and a clear link between a high ratio of omega-6:omega-3 fatty acids and the development of these pathologies.”[3]

    As for arachidonic acid intake, there has been an increase in both meat consumption and the arachidonic acid content of that meat all over the globe. Norwegian animal science researcher Anna Haug discusses the consequences: Many health effects of oils rich in oleic acid (OA, 18:1 n9) seem to be opposite those of arachidonic acid (AA, 20:4 n6), i.e. concerning cardiovascular risk. In recent studies in humans and in the rat we observed that percentages of OA and AA were inversely related, raising the question of whether the inverse association is a general one, and how it might be explained… It is widely accepted that oleic acid, and oleic acid rich foods such as olive oil may have many beneficial health effects. Among such effects are improved insulin sensitivity, and endothelium-dependent flow-mediated vasodilatation, lowering of LDL cholesterol and an increase in HDL cholesterol. If lipids in LDL are enriched in oleic acid, the particles will be less liable to be oxidized, a property that is of significance for the normal metabolism of LDL. Furthermore, intake of oleic acid seems to be associated with reduced blood pressure. Thus, many of the effects of oleic acid may serve to reduce the risk of cardiovascular diseases. Additionally, the fatty acid may have anti-carcinogenic and anti-inflammatory effects… An interaction between oleic acid and arachidonic acid was suggested several decades ago in the rat. More recently, Cicero et al. showed in human subjects that supplementation with a high dose of olive oil for 3 weeks resulted in an increase in LDL oleic acid and a decrease in linoleic and arachidonic acid… One mechanism by which OA could counteract those of AA is to reduce the relative abundance of AA in serum and tissues. Conceivably, increased supply of oleic acid might reduce that of AA by pure mass action. Inverse regulation could also be effected through more specific metabolic feedback regulation. For example, a reduced percentage of AA would be expected if OA inhibits Delta-6 desaturase, Elongase-5 (Elovl-5) and/or Delta-5 desaturase, the enzymes governing formation of AA from LA. Conversely, inhibition by AA of Delta-9 desaturase should lower percentage OA, and previous studies suggest that this latter mechanism might take place… It seems that the Delta-9 desaturases are of considerable physiological significance. Thus, regulation of the amount of monounsaturated fatty acids (MUFA) has the potential to affect a variety of key physiological variables, such as insulin sensitivity, metabolic rate, adiposity, atherosclerosis, cancer and obesity.[4]

    Is the above a biologically plausible explanation for why the Mediterranean diet consistently produces favorable results? It would seem so. Typically, a switch to Mediterranean-style eating reduces linoleic acid intake (olive oil), decreases arachidonic acid intake (reduces meat consumption) and increases omega-3 intake (adds fish).


  4. April 24, 2019 at 10:58 pm, Emily Dhurandhar said:

    Thanks for covering this- interesting talk from Kevin!
    The “what diet is best” conversation has gotten lost in the macronutrient weeds for a long time- which is ironic to me, given that I have never seen evidence to suggest macronutrients are strong drivers of eating behavior and energy intake in a way that transcends their energy value.
    But, palatability certainly does drive eating behavior and intake, and processed = palatable. So, I do see some common ground there.
    I think the challenging next step will be in translating Kevin’s findings regarding processed foods in excess, every meal causing weight gain to a real world setting. Are we really exposed to that much processed food all the time? What about the elements of choice, culture and social norms, stress that we’re absent from the lab setting, just to name a few? In his study all of those factors were taken out of the equation. Also, how representative were those 20 individuals of the eating behavior spectrum? I am eager to see if the effect size he detected applies to more variable contexts and people, and I am hesitant to conclude that simply the presence of processed food would always have that same measurable effect.

  5. April 25, 2019 at 9:24 am, John DiTraglia said:

    I agree with you Emily Dhurander – what he was measuring was probably palatability.
    And Dr Hall when he said, “What this tells me is that the body does a pretty darn good job of adapting. It is more or less as efficient on these different diets. It’s amazing. Evolution made this to be so. We can survive in a variety of different environments, eat a variety of different diets, widely varying in carbohydrates and fats. It’s not magic.”
    The bottom line is wt change over a long time…