Man chewing green apple

Could there be a connection between having to chew food and health? Is “chew your own food” a dictum that will improve your health and life expectancy? Let’s take a look.

The concept that fruit juice is not as healthy as whole fruit, or even not healthy at all, is not a new one, but the more general notion that you should “chew your own food” was suggested to me over on Twitter by Allan Folz, a frequent commenter on this site.

Fruit juice is associated with the metabolic syndrome

It’s well-known that sodas and other sugar-sweetened beverages – SSBs, the term of art used in nutritional science – lead to weight gain and disruption of metabolism. Frequent consumption of fruit juice is also: consumption of more than 5 servings a week of fruit juice, whether natural or bottled, is, along with other SSBs, associated with increased risk of the metabolic syndrome.

The authors of the study speculate that lesser amounts of fruit juice, at less than 5 servings a week, may not be as harmful because of polyphenols in the fruit that may counteract harmful effects of sugar.

Since this study is observational, it can’t show causation. But we know that sugar and SSBs spike blood sugar and insulin, so there’s reason to think that the association could be causal, at least in part.

Matthew Dalby has looked at the science on the metabolic reactions to whole fruit as opposed to fruit juice. Orange juice caused a greater rise in insulin than whole oranges, and the juice caused a reactive hypoglycemia. Results were similar when comparing apples to apple juice.

So, eating whole fruit doesn’t appear to cause the same glucose and insulin responses as does consuming fruit juice.

Dense, acellular carbohydrates

Many people believe that carbohydrates cause obesity, or that they’re a large contributing factor, but a few facts get in the way of that thesis.

Notably, many populations have eaten high amounts of carbohydrates throughout history, and in the present, without suffering high rates of obesity. For example, the Kitavans, so often held up as an example of excellent health, eat a diet that contains about 75% of calories as carbohydrates, and they have no cardiovascular disease, cancer, or obesity.

Mass obesity is a recent phenomenon.

The answer to this question could be that it’s not carbohydrates that cause obesity, but a particular type: dense, acellular carbohydrates.

Comparison with ancestral diets suggests dense acellular carbohydrates promote an inflammatory microbiota, and may be the primary dietary cause of leptin resistance and obesity

The author, Ian Spreadbury, writes that neither glycemic index, fat, or carbohydrates can be causes of obesity, but that flour, sugar, and refined fats alter the gut microbiota, causing inflammation and obesity. Similar to the effects of sugar on dental health, the dense, acellular carbohydrates – flour, sugar, and processed food – cause inflammatory reactions and changes in the gut microbiome that tubers, leaves, and fruit do not.

Processed grains and sugar offer a greater density of food and affect the body differently from whole, unprocessed foods. Adding fat to the mix, especially in the form of seed oils, produces a “double hit”, with the result being mass obesity.

Mechanical disruption of food

It seems likely that mechanically disrupting food, and especially when fiber is removed, changes the way the body handles it. Mechanical disruption, such as grinding of flour and juicing of fruit, creates a calorie-dense food that humans are not adapted to by evolution. Our genes are not meant to handle it.

Nut butters are another example of mechanically disrupted food. I’ve read, but can’t currently find the answer, that nut butters provide more calories than the nuts themselves, simply because the butters have more surface area and the gut therefore digests it more, whereas a good proportion of actual nuts pass through the digestive tract without being digested.

Don’t drink your calories

The adage “don’t drink your calories” comes from a study that found that drinking caloric beverages with a meal, whether the beverage was milk, cola, or orange juice, led to the consumption of more calories while the subjects reported no difference in satiety.

There’s something about liquid calories that doesn’t register with the normal appetite control system, so if you drink calories, that could put you on the road to weight gain, or stall weight loss.

The same principle might apply to seed oils or even olive oil. Or milk or cream.

In fact, using this principle, we may not even need to invoke the ideas above about how the gut microbiome is affected by process foods. It could be that they merely provide more calories with less satiety.

Don’t eat processed food

The big lesson here is: don’t eat processed food.

Almost by definition, it’s been refined and mechanically disrupted. Processed food is likely a big factor in obesity, either because of its refining, including added refined sugar, or because most of it contains seed oils.

Whole, unprocessed foods are not mechanically refined, and don’t appear to upset appetite regulatory systems or the gut microbiome, and do not cause inflammation.

Another point is that carbohydrates may not be so bad as long as they’re in a more natural state, such as vegetables, tubers, fruits, and greens. Although if I were overweight I would certainly be careful even with those. Physically active people may, however, have little problem with them.

PS: Check out my Supplements Buying Guide for Men.

Share this post with your friends

Want to permanently lose 20-50 pounds in 12 weeks?

Without going hungry, counting calories or doing cardio


  1. Rob H says:

    Hi Dennis, very thought-provoking article. Based on all this have you in any way altered your opinion on whey protein shakes? Personally, I take a 40g whey protein isolate shake each morning / lunch to break my overnight fast. I have also recently taken to supplementing my other 2 meals with 3g of leucine to ensure maximum MPS during the day. What are your current thoughts on that? All the best, Rob.

    • P. D. Mangan says:

      Hi Rob, I still take whey protein, but mainly just after workouts. I realize that there may be a few holes in my thesis above; for instance, I eat yogurt, which is kind of half-liquid, half-solid. Some or all of the problem with caloric drinks may be the sugars in them; for instance, the Masai drink cow’s blood and it doesn’t harm them – of course they walk like 100 miles a day too. I think if you want to have a whey shake for breakfast, that’s fine, and will likely help you keep your hard-earned muscle. As for leucine, I still use it in the way you describe on occasion, i.e. when I remember.

  2. David Nystrom says:

    I have an N = 1 caveat to this article.

    For nearly one year nearly my entire caloric intake was in the form of smoothies made in a high powered blender (1500 watt motor). Doesn’t get more mechanically disrupted than that.

    However, the smoothies I consumed contained mostly low or no carbohydrates. Generally each smoothie consisted of partially cooked eggs, whey protein, cream or olive oil, and spices. Early on I incorporated frozen vegetables, but I eliminated these over (probably unfounded) fears of oxalic acid.

    I consumed two of these smoothies most days, and I do not snack. I experienced no weight gain or any other physiological changes.

    Drinking your calories seems fine if it’s a LCHF meal.

    I still wouldn’t advise this regimen to other people (and I stopped myself as I got sick of drinking smoothies despite the convenience) as it probably has detrimental effects on your dental health and jawline. We are, after all, meant to chew.

    • P. D. Mangan says:

      Well, that’s certainly interesting. Did you consume smoothies just for convenience? Also, I agree that sugars in drinks are probably the main culprits.

      • David Nystrom says:

        Primary reason was convenience. Secondary reason was cost savings compared to meat.

        Largely abandoned it because at the end of the day I like food and cooking–I just don’t just cleaning stoves.

  3. brbr says:

    Since meat is relatively expensive, is there some cheap protein, or should i just ditch all the processed meat like hamburguers, sausage, etc…

    There is still eggs i guess, but only eggs takes my motivation away 🙁

    • P. D. Mangan says:

      Some meat is relatively cheap, such as pork and chicken, and I eat a lot of ground beef (hamburgers) myself, and it’s not expensive.

    • David Nystrom says:

      As Mangan noted, pork and chicken are cheap. Bacon is pricey, but salt pork isn’t.

      Beef is more nutrient dense than pork and chicken, and it tastes better. Unfortunately it’s priced accordingly.

      There are some reasonably priced beef cuts however. Ground beef, especially 73/27, tends to be quite cheap. The tougher cuts from the front of the animal such as chuck and brisket as cheap as well. They’re also especially convenient as they’re well suited for slow cooking or stewing in bulk, allowing you to prepare a very large number of tasty meals at once (whereas leftover steak sucks).

      You can liven up your eggs with cured meats and aromatics.

      And while I don’t eat them myself, legumes have undergone a partial rehabilitation in the paleo world in the past few years. Be sure to soak legumes overnight prior to cooking.

      Lastly, sign up for and look for meat and seafood sales in your area.

    • TeeDee says:

      As P.D and David N. mentioned, ground beef is reasonably priced and more nutritionally dense. It can be made into your own tasty burgers, esp. when ‘smashed’ flat and seasoned well before frying. Or it can be made into a bean-less chili, or mixed with mushrooms, onions and cream if you do dairy now and then.
      I was actually stunned to realize that my once-in-a-while treat of chicken wings baked with just salt, granulated garlic and cayenne pepper is more expensive than lean ground beef patties. As for sausages, I buy fresh pork sausages from an Italian butcher that contains nothing but pork, salt, pepper and garlic. They work out to just over $4.00 per lb and are versatile as well as very tasty. All the best.

      • TeeDee says:

        Oops. I should mention that the $4.00 per lb mentioned is in Canadian dollars, so they would likely be less in U.S. dollars.

  4. Bill says:

    PD, Thanks for this post.. Made me think again and reminded me of the basics yet again !

    But I wonder about one aspect you have not touched on : genetic variability among us humans.. Could it be that after a thousand years or so of relative isolation Kitavans are now genetically adapted to their 75% high carb diet ? And maybe some individuals are also genetically better adapted to a higher carb diet ?

    PS No argument at all about high sugar & industrial fats in processed foods. etc..I stopped drinking even organic juice a few years ago. Avoidance is the only way with these items

    • P. D. Mangan says:

      Bill – Certainly genetic variation could be a factor, but (IMO) it’s an overrated one. One thing about the Kitavans, as well as the Okinawan centenarians, is that they are short, and short people live longer. Same with people in the Mediterranean.

  5. Robert says:

    Hi Dennis,

    What about blended veggie shakes? I recently started doing this to get some extra fibrous veggies in my diet. I specifically opted for a blender over a juicer to keep the fiber and slow down the glycemic response.

    Im still downing the shakes in a few minutes as opposed to slowly chewing and eating smaller portions of cooked veggies over the course of 10 minutes. Do you think its worth reconsideration?

    • P. D. Mangan says:

      Robert – I suppose that’s a fairly low-calorie drink and low in sugar, but ultimately my answer is I don’t know.

  6. bigmyc says:

    I think that the key mechanism for the insidious weight gain due to the consumption of juices and other liquid sustenance is how readily absorbed they are…in some cases, just about immediately. This can’t be good for one’s insulin status and if there’s one indicator of overall metabolic health, it’s the circulating insulin count and the degree and frequency of insulin spikes.
    Carbs are not “evil,” but it just pays to keep them to a minimum simply because doing so keeps the more pernicious items in the carb family (gluten, lectins, excess omega 6 fatty acids, etc.) out of one’s diet.
    Though they aren’t in great supply, there are carbs in fruit and vegetables. However, there is also copious amounts of fiber which slow the rate of glucose intake into the metabolic system. Even potatoes have a fibrous element to them that juices don’t have. Other than pectin, there is no fiber in fruit juice.

    Altogether though, I don’t think that the “liquid principle” is all encompassing. I don’t see the same mechanism with milk and cream and certainly not olive oil. Again, the glucose uptake isn’t going to be so pronounced, even with the milk, because of the fact that the fat content will ameliorate that.

    • Rob H says:

      I’m not so sure that insulin spikes are a bad thing. I was recently reading an article on Lyle McDonald’s site and I came across the following which was a real eye-opener for me:

      “A long while back, I described a study where a low GI food actually caused a larger EARLIER insulin response than a high GI food. If this seems confusing remember what insulin does which is to push glucose into tissues such as muscle. The blood glucose response is ultimately related to how much glucose is exiting the bloodstream as well as how much is entering. If you push glucose out of the bloodstream faster than it is coming in, you get a lower measured GI response. But it’s because insulin is going more quickly to a higher level.
      Similarly, I noted that adding protein to carbohydrate invariably lowered the measured GI response. And it turns out that it was because adding protein to carbohydrate INCREASES the insulin response. This means that blood glucose is pushed out of the bloodstream more quickly. The measured GI response is lower but it’s because insulin is going up MORE. Even in the bran cereal study linked above, the researchers thought that the increased insulin response was due to the higher protein content of the bran.”

      I thought this was a pretty powerful insight that I was not aware of: ie that lower GI foods are lower GI precisely because they elicit a greater and more rapid spike in insulin which then proceeds to clear the glucose from the bloodstream more quickly. For similar reasons, I feel that the high insulin spikes caused by the branched chain amino acids in whey/ dairy are not in fact a bad thing – they are in fact highly beneficial in that the elevated insulin shuttles the aminos to the muscle more quickly. Apparently the insulin ‘area under the curve’ then falls rapidly. So, I see rapid spikes in insulin as generally a good thing (for healthy individuals that is) – it is chronically elevated insulin that is to be avoided. Were you aware of this study Dennis – what are your thoughts on this?

      • bigmyc says:

        Sure…albeit in an extremely acute condition. Allow those massive insulin spikes to continue, eventually you wind up with insulin resistance.

        Insulin spikes might indeed be “good”‘ for those who have otherwise achieved optimal metabolic health, but those who are “metabolically deranged” don’t need any more circulating insulin than they are already chronically exposed toward.

        • Rob H says:

          Hi bigmyc, what you say in terms of large insulin spikes only being good for those who have achieved optimal metabolic health seems at first glance to make intuitive sense but I just don’t think it holds up when you dig down into it: for example here is a quote from Mark’s Daily Apple ( that clearly shows that large insulin spikes relating to whey are beneficial for both non-diabetics and diabetics alike:

          “Diabetes – Eaten before a meal, whey reduces the glucose spike from the subsequent meal in non-diabetics and type 2 diabetics alike. It achieves this by “spiking” insulin, but transiently; the insulin area under the curve improves even as the immediate insulin response increases. Plus, as seen above, fasting insulin tends to lower in people consuming whey protein.”

          The links to the studies don’t show up in the excerpt above, but you can see them via the link to Mark’s Daily Apple above. So I maintain my point of view that large insulin spikes can be beneficial to one and all – in the specific case where cells are nutritionally depleted (eg depleted glycogen/ muscular amino acids etc) and where the insulin spike quickly falls back to baseline (ie insulin is not chronically elevated). I do this myself at least once a day, sometimes twice using whey protein isolate shakes and BCAAs straight after finishing an intermittent fast or HIT exercise (or both) – and I believe these large and regular insulin spikes to be beneficial. I’d be interested though if you can show any evidence to the contrary?

          • bigmyc says:

            You are still (as is Mark’s Daily Apple) referring toward an acute situation with the insulin spikes. These are “manufactured” incidents to facilitate a positive end…
            ..if you recall, I originally cited ” if there’s one indicator of overall metabolic health, it’s the circulating insulin count and the degree and frequency of insulin spikes.”… of course, I’m talking about the insulin status that manifests itself upon a natural, everyday level; that status which one’s metabolism defaults toward on a chronic basis. High Insulin is never something that anyone, diabetic or otherwise, should aspire toward. I’m sure that you’d agree with that.
            However, if the insulin spikes in question are deliberately administered, isn’t that of a quite different nature than if the spikes were endemic to one’s metabolic operations?…..

          • Rob H says:

            Hi Bigmyc, yes I think we are pretty much on the same page here – we are probably just debating semantics. I’d definitely agree that as you say” high insulin is not something to aspire towards” – so long as you are referring to high CHRONIC insulin, ie significant increases which are sustained over a course of several hours – ie serving no positive functional purpose. But, where you mention “the degree and frequency of insulin spikes” I am inferring you believe the higher the spike and the more frequent the spike the worse off a person will be? At least, that’s how I read your comment or did you not mean it that way? I still believe (and the links in the MDA piece seem to bear it out) that even for diabetics, a greater frequency and amplitude of insulin spikes is a beneficial state to be in – regardless of how these spikes were administered. The key point is that the spikes must be short-lived with insulin returning to baseline fairly quickly: ie an ACUTE response – not a chronic response, and given in a nutrient deplete scenario – so that the insulin spike can fulfill it’s intended purpose: ie swiftly delivering nutrients (glucose/ amino acids) out of the blood and into the cells that need to replete themselves. Anyway, just my thoughts – sounds like we are pretty much in alignment here!

          • bigmyc says:

            Pretty much but you still need to consider how metabolic derangement occurs; the consistent (chronic) elevation of insulin. So…frequency and not so much degree, is what I’m citing here. Of course the removal of glucose (and nutrients et al) is a beneficial thing, after all, it’s why the pancreas exists…but let’s remember, what we’ve both agreed to so far; insulin tolerance is not a good thing. I hope we aren’t splitting hairs at this point, but this part about “short-lived with insulin returning to baseline fairly quickly,” wasn’t ever mentioned. I would maintain that this is elemental and forgone in relation to the sum total pernicious effect of excess insulin circulating throughout the pulmonary system.

          • Rob H says:

            Yep, I think you’ve pretty much covered it there bigmyc. The whole area of insulin control is of great interest to me (as you may have noticed!) together with ways we can manipulate it ourselves via fasting, HIT resistance training, cold showers and also ingesting highly processed proteins such as whey protein isolate and branched chain aminos, in particular leucine which seems to be the key player with regards to stimulating insulin spikes. Dennis, not sure if you’re seeing this, but would love to see any more you can unearth on these areas!

            So just to summarise my position (where I am defining ‘spike’ as something which reaches a high peak very rapidly then falls rapidly, say within an hour or so):

            High chronic (continually raised) blood sugar: BAD
            High acute blood sugar spikes: BAD
            High chronic (continually raised) insulin: BAD
            High acute insulin spikes: GOOD

            Where the high acute insulin spikes actually cause the lowering of blood sugar, both acutely and chronically.

            Think I’ve said my piece on this subject for now!

  7. Allan Folz says:

    Thank you, Dennis. I’m basking in my 5 minutes of Internet fame.

    If I may be so bold as to underscore a point, the study at the apples link is what really caught my eye. It was a small experiment done back in the early 1970’s. There was some poorly controlled research back then, but on the other hand, it was largely before all the perverse incentives we have today. Research was much more honest, even when it was shoddy. As for me, I’ll take shoddy over subterfuge every time, thank you very much.

    Anyway, they compared apple juice, apple smoothies, and whole apples in a kind of mini-Kraft test. They kept everything constant they could think of, including time to ingest. The only thing not controlled was, obviously, the surface area of the ingested apple. Sure enough, whole apples chewed-up had an entirely reasonable insulin response; smoothies, not so much; apple juice, none at all.

    There’s been a lot of digital ink spilled on fiber. In fact, I’m rather sympathetic to the idea that we need to keep our gut biome healthy and strong. However, as far as insulln response, which we absolutely know to be important, there’s no question, surface area matters.

Leave a Reply

Your email address will not be published.