Kurt G. Harris MD

PāNu means paleonutrition. The "paleo" here signifies "old" and not necessarily paleolithic. The PāNu approach to nutrition is grounded on clinical medicine and basic sciences disciplined by knowledge of evolutionary biology and paleoanthropology. The best evidence from multiple disciplines supports eating a pastoral (animal-based) diet rather than a grain-based agricultural one, while avoiding what I call the neolithic agents of disease - wheat, excess fructose and excess linoleic acid.

Support PāNu

PāNu is ad-free, completely independent and has no outside sponsorship. If you value PāNu, now you can support it. Read this for more information.

In addition to buying from the book list, you can also support PāNu by making all of your Amazon purchases for any item through the Amazon Portal below

Amazon Portal

« Smoking Candy Cigarettes | Main | A Taste of Dairy »

Insulinogenic is not Hyperglycemic

Reader Dee writes in with her experience as a Type I diabetic. She has some good observations on what a large protein meal does without endogenous insulin.

There is a lot of mention of dairy being insulinogenic. Anecdotally, this is what I have seen as a type 1 diabetic. Beef causes my blood sugars to rise very high about 7-8 hours post prandial. So does cheese - in similar quantities. Butter and cream and full-fat yogurt do not. But the difference is a serving of cheese might be 1-3 oz. A serving of beef on a low carb or paleo diet might be 8-16 oz. So eating small amounts of cheese are not a problem, insulin-wise. However eating large portions of beef in particular, and other meats as well, require more insulin than 30g of carb - it's just much later on. In an attempt to lose weight I need to keep my insulin dosages low - therefore I must limit my protein. Ditto on carbs. If I don't eat more fat and calories I am worried about stunting my metabolism. It seems very difficult to up your fat intake without dairy (cream, butter). That's just my pragmatic rationalization as to why I'm eating dairy. I do try to choose organic cream and pastured butter to minimize the hormonal gunk though.

She has shown nicely how large protein meals require more insulin or her blood sugar will rise.

She also got me to thinking about how I see the term "hyperinsulinemia" used at times.

In the blogosphere, I occasionally see some conflation of the terms hyperglycemia and hyperinsulinemia. On a chronic basis, they are part of the same pathology of metabolic syndrome, but as part of normal homeostasis they are not.

Insulinogenic means "Induces an insulin response".

Rising blood sugar does not automatically equate to rising insulin -in fact, as any Type I knows all too well, hypergyclemia is due to inadequate insulin response. Your serum blood glucose (BG) can rise in the presence of an insulin response or without it - it depends on glucagon and epinephrine and other foods eaten as well.

It is of course true that there is an insulin response to amino acids- and this is roughly an order of magnitude smaller than the response to glucose on a molar basis. It is also true that if more glucagon is released relative to insulin with the protein meal, that serum BG can rise even with the insulin increasing.

If you are type I, your BG is rising due to the lack of an increase in insulin in the presence of increasing glucagon liberating glucose from your liver, not due to the "insulinogenic" effect of 16 oz of beef. In fact, if there were not glucagon released after eating protein, a non-diabetic could become hypoglycemic from the insulin response to the protein.

When Cordain complains that dairy protein is "insulinogenic", he means (or should mean) that a given amount of dairy protein raises insulin more than an equimolar amount of, say beef protein.

The question is not whether this occurs, but if it a large enough effect to be clinically significant.


Thank You, Dee

Reader Comments (22)

The perfect anti-insulinogenic and anti-hyperglycemic diet:

Eat nothing but butter?

KGH: except for those pesky amino acid and brain glucose needs, that would be the ticket.

January 11, 2010 | Unregistered CommenterJim Purdy

@reader dee: not only insulin levels determines if you will lose weight. despite it's important for free fatty acids use, dont forget that a calorie deficit is the most important variable.



Which comes first, the caloric deficit or the hormonal drive? It is not more important than the hormonal milieu and unless you are stuffing yourself, energy balance is mostly a consequence of it, not a separate parameter.

Have you not read Good Calories, bad calories?

January 11, 2010 | Unregistered Commentersoultransfer

A great article, reinforcing the notion around here about "let protein come along for the ride"

KGH: Exactly. I push animal fats, but don't measure protein as tightly as Optimal Diet. I reckon with a healthy metabolism and avoidance of the neolithic agents that 5 -30% carbs and 15-30% protein are reasonable. Note that is what you get if you try to control only fat and add some veggies to it. Fat in the range of 50-80% is the key PaNu parameter, IMO.

High sat fat is the anti-atherogenic anti-degenerative nexus. More on that later.

This makes PaNu convergent with Protein Power (Eades), Atkins and close to K's Optimal Diet as far as Macro ratios.

So PaNu is the big tent for Macro ratios, but the Neolithic agent exclusions are stricter than the above approaches, yet not as strict as the more "ideologically pure historical paleolithic" diet like Cordain's

January 11, 2010 | Unregistered CommenterScott

Contrary to stereotype, there is a (significant?) minority of type 1 diabetics who are insulin resistant and have weight issues. From what I have seen personally, and around the web at diabetic communities, it is very difficult for these overweight type 1 diabetics who want to lose weight to do so. I think it may have to do with a post that Peter made at Hyperlipid that indicated that exogenous insulin does not effectively signal the pancreas. Also, there are other hormones that type 1 diabetics lose with their pancreatic function besides insulin - amylin comes to mind. Well, that's why I read these blogs - to try and figure it out.

With regard to calorie defiicit - stopping insulin altogether is not a great option - but does produce results, which is why some type 1 teenage girls in particular take this approach for weight loss and why one of the characteristics for diabetes diagnosis is rapid weight loss.

Anyway, I think I understand the distinction you are making with insulinogenic vs hyperinsulemic, however I have always viewed it as in a healthy body, these substances are stimulating insulin to be produced because the insulin is needed to deal with some sort of glucose production. And for me, it's a DIY project. But are you saying that it is stimulating unnecessary insulin in the body that then needs to be dealt with via counter-regulatory measures? I guess if something is insulinogenic but doesn't cause a rise in blood sugars, it is excess and unneeded insulin production.


"Insulinogenic" is often used a pejorative but we should be careful to understand (as you do all too well) that both insulin and glucose are necessary to life. The point is to avoid chronic hyperinsulinemia whether normal or diabetic.

Bernstein covers this well and that is why paying attention to protein matters for diabetics who inject insulin. Generally using less injected insulin is the analogue of avoiding metabolic syndrome in non-daibetics.

Whether avoiding a little extra protein matters much in those with normal metabolism ( as Cordain and Kwasniewski would say) is somewhat debatable, IMO.

January 11, 2010 | Unregistered Commenterdee

Hi Dr. Harris,
I apologize for the second question today but this one still bugs me. Coming from the Paleo camp, they talk a lot about the acid/base balance which is why they advocate eating lots of vegetables. You might have talked about this before but does this positioning have any scientific merit? I just have very little desire for veggies beyond a salad or some asparagus.
Thanks as always.


Here is the honesty that I frequently brag about even if it attenuates my Sensei status.

I don't know.

This is another one of Cordain's hobby horses, as well as, ironically, being an an established trope in the "meat is unhealthy" armamentarium of the vegetarian mythology.

I've certainly read papers on the alleged mechanisms of how failing to balance your meats and dairy with vegetables or eating too much protein can "leach calcium from your bones" but so far I don't find it very convincing.

The idea that we evolved to be intolerant of a parametric range of acid/base balance in our food, and that HG tribes of old easily got osteoporosis if salads were scarce strikes me as a bit implausible.

January 11, 2010 | Unregistered CommenterMark

I would tend to agree. I feel great on butter, lots of cream, beef, lamb, pork, some chicken, some veggies, some rice or sweet potatoes, little fruit and honey, some nuts, and lots of eggs. Thank you for helping me know that this approach is healthy.

January 11, 2010 | Unregistered CommenterMark

Is it also possible that there's something in between normal and diabetic wrt protein and insulin? For example, I just wonder if some of the non-diabetic folks who plateau on Atkins or other low-carb diets are running into problems from an insulin response due to excess protein in the diet being converted to glucose.

As an aside, what do folks think about Nora Gedgaudas' case for moderating protein because of its impact on our mTOR pathway?


I don't get the science entirely, but find it interesting. And she also mentions that "'carbovores' who try to switch to eating 'high protein diets' may be just that much more efficient, too, at turning that excess protein into sugar and storing it the same way."

To me (I know, a scary lay person :), it suggests that for people with problematic insulin metabolisms (not just diabetics), moderating protein may well be a good thing.




Enough quoting Cordain -I disagree that it is at all established that this is a cause of clinical cause of chronic hyperinsulinemia. I simply do not have time to rebut every misleading Cordain quote line by line.

I have left the link to the article intact for those who want to read it.

All those amino acids are also found in meat.

Gedgaudas' book I have only skimmed but it contains a lot of speculation.

January 11, 2010 | Unregistered CommenterBeth@WeightMaven

I'm entirely with you on the question on rebutting Cordain. I just included it because of your comment in the original post questioning what Cordain meant by "insulinogenic".

Me, I'm just trying to figure this out because I'm not like you or Stephan Guyenet ... someone who has the luxury of currently being in a fairly good state of health for whom these details are more or less academic. Kinda of a bummer that all of this sounds pretty damn convincing to a lay person like me. Anyways, thanks way much for your blog. It really rocks!


Thanks Beth - I appreciate your support and enthusiasm. I don't have diabetes but I try to take all this as seriously as if I did. I'm older and probably more SAD - damaged than Stephan :)

January 11, 2010 | Unregistered CommenterBeth@WeightMaven

Kurt, with your recommended diet, how do you solve the problem of high insulin levels due to the high iron accumulation due to the high red-meat diet you recommend?

KGH Whenever someone quotes me as recommending a "high red meat diet" I suspect vegetarianism - they are the only ones who think butter and cream and egg yolks are red meat :)

Like in this study it found that drawing blood to deplete one group's iron levels was more effective at lowering insulin levels than a second group that tried to control insulin just through a low-glycemic diet:


KGH: That study obviously concerns people that have a serious disease.

And this study here shows that vegetarians have lower insulin levels than omnivores mostly because of their low iron stores:


Do you think the best way to do it is to eat a high meat diet but also donate blood to keep the iron stores in check? What about other things that would accumulate heavily on this diet, like calcium for example?

Thank you for your opinion.


I've heard the "donate blood" thing before. My ferritin levels are mid-range normal and I don't donate blood.

Please remember that it is liver insulin sensitivity specifically that is related to met syn, not just a crude glucose challenge. Vegetarians have to have better glucose uptake than VLC eaters to avoid diabetes with all the carbs they eat! VLC eaters have more muscle insulin resistance to avoid passing out from hypoglycemia. Read Peter's posts on physiologic insulin resistance.

Higher whole body (liver plus muscle) insulin resistance is an adaptation to fat fueled metabolism.


January 12, 2010 | Unregistered CommenterMatt

According to Holt (and Brand Miller) Am J Clin Nutr 1997;66:1264-76 the insulin AUC over 2 hours in healthy people after equal calorie servings of cheese, beef, pasta, rice and fries are all very similar although the glucose responses are different. Dont know what happens after 2 hours. My view would be that it confirms insulin is an anabolic hormone rather than a sugar hormone and it is probably glucose levels which drive the metabolic problems not insulin levels. I dont know whether gluconeogenesis is a problem or not but my prediabetic blood sugars are better if I moderate protein and do not over eat in general.

KGH: Got other posts to write now - may comment later - next time give a link to full text if you want me to comment.

January 12, 2010 | Unregistered CommenterMichael

Dr. Harris,

Thanks for a truly great blog!

Quick question: I take it you make a point of eating pastured butter. May I ask what kind/brand, and where you source it?

Also, do you eat grassfed beef? If so, any sources you like?

Thanks again.

KGH: My grassfed beef and lamb are sourced locally. My pastured butter is Pastureland organic unsalted - a cooperative from Minnesota

January 12, 2010 | Unregistered CommenterTom


Thanks for an enlightening post. This is quite helpful in deciphering some of my own lab values, as well as those of Lex Rooker, who has commented on your posts in the past.

I have irritable bowel syndrome (IBS). The best way I found to control it involved having Lutz-like ~75 carb grams/day (though I was using neo-, not paleo-carbs). But I still had symptoms, so I tried reducing my carbs more. I still had symptoms so, thinking my remaining symptoms were caused by the remaining carbs, I tried going zero carb (ZC).

After 6 months of ZC my HbA1c was 5.9%. I was surprised to find it this high. And after 7 months of ZC, I sometimes had fasting BG around 101. Again, I found this surprising. Most importantly, my IBS symptoms worsened, so I abandoned low-carb.

I've had 4-5 grams carbs/day for the last 5 days, and my BG is, if anything, lower than it was on ZC. This reinforces your statement that "insulinogenic" need not be used only pejoratively.


4-5 g a day is not much of course, and would only measurably affect your BG if you were diabetic.

The idea that you can achieve arbitrarily low levels of blood glucose by eliminating carb consumption is one of the more nonsensical ZeroCarb notions. Your body will work to maintain serum BG for benefit of your brain. M

January 12, 2010 | Unregistered Commentermoises

After a year plus of mainly carnivorous eating and endless blog/forum reading you have been the sole person to mention bovine serum albumin allergy and you've mentioned it a few times. If you have the time, I'd like to hear more of your thoughts on it. Thanks

KGH: I mention it when folks maintain that allergy to casein proves we are not meant to eat dairy products. Does allergy to beef prove we should never eat beef?

January 12, 2010 | Unregistered CommenterTeddy

Dr. Harris,

Have you seen or read this paper? http://pcwww.liv.ac.uk/~gowlett/GowlettCJNE_13_03_02.pdf

Someone sent this to me and I'd like to hear your thoughts when you get the chance.



KGH: I'll add it to the pile -no guarantees.

January 12, 2010 | Unregistered CommenterAl

Dr. Kurt:
You said::
"It is also true that if more glucagon is released relative to insulin with the protein meal, that serum BG can rise even with the insulin increasing."

Wow, that set off a light bulb for me. This means there can't be pat assumptions on what is really causing higher glucose readings/HBA1C on VLC/ZC diets, right? It might be gluconeogenesis OR glucagon unopposed by enough insulin causing too much BG to be released from glycogen (any other causes?). Most people seem to be only focusing on gluconeogenesis.

Thanks for all your stimulating posts!

KGH: Precisely. Protein meals stimulate glucagon - they don't instantly turn into glucose!

January 12, 2010 | Unregistered CommenterJanet Gomez

i do not intend to argue about every single point. and to be honest i'm not much into that and do not have a real opinion about that.
i just know, that it was easy for me to gain weight with a lot of butter and chicken (with skin) without much carbs. so at the end of the day calories matter. it's the single variable that affects weight loss or weight gain.
you can eat 80% carbs, 0% protein, 0% fat. as long as you are in a calorie deficit you will lose weight. no matter how you hormonal millieu looks like.
that this way of eating will set your hormonal system up and makes it easy to overeat is another issue.
would you agree?

so that was my thought that came to my mind as i read dee's quote.
"In an attempt to lose weight I need to keep my insulin dosages low "
and i want to respond.
YES, AND YOU NEED TO GET IN A CALORIE DEFICIT. dont just focus on insulin levels.

and i thought about buying garry taube's good calories and bad calories. so i did some research at lyle mcdonald's forum. if it comes to weight loss, there is no one who gives better information.
stumbled accross this thread where he gives his opinion about gcbc:
i would like to give you the url


i'm really interested what your thoughts are.

KGH: Why don't you actually read Taubes yourself two or three times (seriously, so it sinks in) then decide for yourself if caloric intake leads hormones or follows. Pullout some biochemistry texts and read some of his hundreds of references along the way. Then read LM and see what you think of his opinion. I know who I put more stock in.

January 12, 2010 | Unregistered Commentersoultransfer

i never denied that caloric intake follows hormones. pls dont mix up my words. maybe i will read garry taubes book one day.

KGH:It is Gary Taubes and you should read it before arguing with folks about insulin or protein on the internet.

i read dee's post and saw someone who is in fear of any insulin response. beside the fact that the insulin response of protein is interesting, it's not the determining variable for weight loss. she is not able to eat beef because of the insulin response, come on!

KGH: No, she is not able to eat beef as much as a normal person because she has DIABETES and has a defective insulin response - it is not the insulin response, it is the glucagon.

set your protein intakte maybe 1g/kg, set your carbs maybe <50g, fill the rest up with fat ...amount depending on caloric deficit.
would you pls tell me if you agree, KGH.

KGH: If you are saying most people need to tightly control macro ratios , esp. protien, I do not agree. The point of 1 g/kg/day, etc. is that is what you need, not that is all you can handle.

January 12, 2010 | Unregistered Commentersoultransfer

According to Bernstein, the pancreas "thinks" BG should be below about 84 mg/dl, because when it does go below that level, the pancreas stops putting out insulin above baseline levels. So, as long as BG is above that level, there is more insulin in circulation than there would otherwise be; but below that level there isn't less insulin, hence the term "baseline."

We care about the "area under the curve" when it comes to insulin because insulin, in addition to pushing glucose into storage, acts as a kind of master controller of other biochemical pathways. Notably, it tilts things toward the production of the pro-thrmobotic, pro-inflammatory series 2 prostaglandins. Barry Sears deserves credit for bringing some degree of popular attention to this, even if it tends to get lost in the blocks and ratios. So, the greater the area under the insulin curve, the more our prostaglandin axis is tilted in a direction that we probably don't want.

On the face of it, then, we have good reason to strive for fasting BG below 84, since we spend a lot more time in the course of a day not eating than we do eating. This, I believe, is precisely the position that Bernstein takes.

But it seems to me that it's a little more complicated than that. Insulin isn't the only knob on the prostaglandin machine. Glucagon is another, and it tilts prostaglandin production the other way, toward anti-thrombotic, anti-inflammatory effects. As has already been pointed out, protein stimulates the production of insulin and glucagon, whereas dietary carbohydrate has little or no effect on glucagon. This means that the effects of insulin after a protein-rich meal should be, to some extent, counteracted by glucagon. So we have less reason to be concerned about the fact that protein is insulinogenic.

But what about fasting BG? If it's elevated, despite low or even zero carb intake, what's going on? It's glucagon, after all, that tells the liver to put glucose into circulation. Does a higher FBG than 84 suggest that glucagon is also higher than baseline? I don't know the answer, but if so, then maybe the elevated FBG is less of a concern. Maybe the significance of FBG depends upon the dietary context, especially carb intake. And maybe one of the questions we need to ask about dairy protein is whether its effect on glucagon is comparable to its effect on insulin.

January 12, 2010 | Unregistered CommenterTodd

Overall, calories still do count. One just cant get around it.

KGH: They count but not in simple minded the calories in calories out sense.

If I eat 300 g of fat in a meal, the fat has to go somewhere. Is it not the fat cells?

KGH: If net fat storage occurs as disequilibrium between hormones and the hormones differ based on equicaloric meals with different content, then you have to have the possibility that consumed fat is not always stored at the same rate.

So there must be something else regulating fat cells to take up fat other than insulin. Also, then there is the fact that Colpo in his book the fat loss bible has listed more than two dozens of tightly controlled metabolic ward studies showing that calories count under various macronutrient ratios, basically there is no metabolic advantage. I mean, you eat carbs, insulin goes up, glucose gets converted to fat and then you can use that fat. Eat it directly or get it converted, whats the big difference?

KGH: All the difference in the world, unless you don't care what happens to your liver or glycating proteins along the way. Macro effects on the hormones mediate fat storage - fat storage drive hunger. Colpo is wrong that there is no metabolic advantage - those who quote laws of thermodynamics show they understand neither physics not biology.

But I do admit low carb diet can be effective by allowing your natural systems to control your hunger based on fat mass, eg making you eat less to get full.

KGH: And how do you suppose it does that - perhaps through an effect on hormones? So hormones are driving both fat storage and hunger.

January 13, 2010 | Unregistered CommenterMike

If you haven't done so already, check out Don Matesz' post on hypoglycemia and polio:
Paleodiet and Polio Virus

The bottom line seems to be that getting glucose down much below 100 significantly increases your chances of contracting polio, at least if you're a monkey or a rabbit.

Just one more thing to wonder about.

January 15, 2010 | Unregistered CommenterWillis Morse

I concluded that the reason I gained 10 pounds going ZC was that I ate pretty much unlimited protein and not so much fat. I have now cut the protein down to Dr. K's levels (counting first, guesstimating later) and gorge on cream and butter.

I eat fat all day and protein once a day.

I have also found that the protein meal puts me into a state of anxiety a couple of hours later. I guess this is the metabolic reaction. It helps to eat a little carb with the protein (about 7g for me).

It is very hard to get enough animal fat if you are on the road and eating at restaurants. The VLC temptation is to take in too much protein.

I have to stop at the grocery store for a package of cream cheese and a half pint of cream per day and keep them in the fridge or on ice at the motel. Or take a thermos for the cream and an ice pack for the cream cheese.

In any case, this is to lend a "shout out" to the idea that protein raises insulin levels and causes weight gain.

They don't tell you this. You have to find out for yourself.

January 16, 2010 | Unregistered CommenterJane

When it comes to calories in, calories out, I am on your side Dr. Harris. I have read Taubes and it is my all time favorite. I recorded what I ate for 5 days. In that five days I averaged over 2500 calories and I didn't have 3 meals (two lunches and 1 dinner). I had a fast on day 3 so day 4 I took in over 4200 calories (all is according to ww.fitday.com). I worked out twice that week (doing crossfit so my workouts weren't more than 20 minutes). About 5 days before I started my recording food I was 208.5. I put on a few pounds after Christmas, porbably water weight from the starchy SAD foods I ate. By the morning of day 4 (and 5) I was 200.5. I lost 8lbs in 9 days. Calories in, calories out is garbage. It is driven by HORMONES!

KGH: I agree.

January 28, 2010 | Unregistered CommenterJeromie
Comments for this entry have been disabled. Additional comments may not be added to this entry at this time.