The “Insulin Index”

Similar to the glycemic index, which is an estimate of the rise in blood glucose after eating a particular food, the insulin index is an estimate of the rise in insulin after eating a particular food.  In general, these indices are obvious: processed carbs have high glycemic and insulin indices, whereas whole foods are lower.  Some exceptions are things like dairy and lean meat, which induce more insulin than you’d expect given to their low carbohydrate content…


When some protein-rich foods were discovered to induce insulin secretion, people thought this information might help type 1 diabetics more accurately calculate their insulin dose.  Interesting rationale, worth testing.

Tl;dr: it didn’t work very well.

More of the protein-derived amino acids may have been incorporated into lean tissue, but the extra insulin load ended up causing hypoglycemia more often than not.  Hypoglycemia is acutely more harmful than hyperglycemia, and is still quite harmful in the long-term.  Some studies on incorporating the insulin index for type 1 diabetics are mixed, ie, increased or no change in risk of hypoglycemia, but no studies show it reduces the risk.



Eg, by increasing insulin dose to accommodate dietary protein, risk of hypoglycemia increased from 33% to 48% in this study.


And from 9.5% to 35.7% here.


Some type 1’s still do this, and that’s fine.  Ymmv.



Part 2.

Drink a little too much of the Taubes Kool-Aid and you might figure that if processed carbs were bad because of insulin, and some protein-rich foods also induce insulin, then said protein-rich foods should also be avoided.

I’ve said there are some holes in Taubes’ theory, and while he never said to avoid protein-rich foods, this may be another one of those holes.  Because just about every single diet study comparing higher to lower protein diets show a benefit of the former.  I also don’t put too much stock in the Rosedale-mTOR theory for the same reason.  Maybe some amino acids activate mTOR, and mTOR over-activation is associated with aging in some animal models, but none of this is reflected in the human studies.  Maybe it’s because protein-rich foods are very nutrient-dense and activate other pathways which nullify any negative effects of insulin/mTOR, or maybe the theory is just flawed.  Or both.


But condemning red meat and dairy because insulin is simply flawed.  You may have other reasons for avoiding those foods, and that’s fine.  It might help some type 1 diabetics more finely tune their insulin dose; that’s fine, too (as long as it doesn’t induce hypoglycemia).



Hypocaloric high protein spares lean mass relative to high carb (Piatti et al., 1994)

Protein: 15% -> 30%: increased leptin sensitivity, reduced appetite, and weight loss (Weigle et al., 2005).


High fat and protein (Atkins and Zone) beat high carb (McAuley et al., 2005).


Dietary protein, ketosis, and appetite control.


If you’re one who benefits from [insulin-suppressive] low carb diets, then keep in mind that there may be more reasons than the insulin-suppressive nature of the diet; protein does not need to be restricted for this purpose (insulin levels actually even decline on higher protein diets in many studies).

For full access to all articles and much more (or if you just like what I do and want to support it), become a Patron! Five bucks a month for full access and there are many other options. It’s ad-free and you can cancel if it sucks ????

Also, I’m open to suggestions so feel free to leave a comment or email me directly at

Affiliate discounts: if you’re still looking for a pair of hot blue blockers, Carbonshade is offering 15% off with the coupon code LAGAKOS and Spectra479 is offering 15% off HERETrueDark is running a pretty big sale HEREIf you have no idea what I’m talking about, read this then this.

20% off some delish stocks and broths from Kettle and Fire HERE

If you want the benefits of  ‘shrooms but don’t like eating them, Real Mushrooms makes great extracts. 10% off with coupon code LAGAKOS. I recommend Lion’s Mane for the brain and Reishi for everything else.

calories proper


Be Sociable, Share!
  • I like to call it Brotein. By the way Bill, do you even lift?

  • High Protein Diets And Body Composition

  • Marty Kendall

    Interesting article Bill.

    The insulin index data would have you minimise carbs and
    protein (to a lesser degree) to minimise insulin. However consideration of nutrient density drives protein and non-starchy veggies back the other way.

    I think there is a balance point for each individual – maximising
    nutrition while keeping the glucose / insulin load of your diet within your own pancreas’ ability to keep up.

    Timing is everything when it comes to insulin dosing for
    protein for people with type 1 diabetes.

    The rise in blood glucose due to protein seems to be delayed
    in type 1s who have only had the disease for a few years and still may have some pancreatic function. In people who
    have had the disease for long protein seems to raise blood glucose sooner meaning that the insulin needs to be given with the meal. The ideal approach usually comes down to
    trial and error for each individual refined using a continuous glucose monitor.

    One dietary approach that seems to work well for type 1s is
    to eat a moderate to high amount of protein with minimal carbs (as per Dr Bernstein). The protein also provides any
    glucose they need via gluconeogenesis while the blood glucose rise is much gentler and easier to manage.

    See this article for a discussion on calculating insulin dose
    and timing for carbs and protein

  • George

    My two cents worth – trying to leverage protein away from your natural preference or “set point” just makes following a diet harder.

  • NY
  • NY

    Does the rise in insulin due to protein inhibit lypolysis in the same way as does the rise in insulin due to carbs? If different people respond differently to carbs why would it not be the case about fat and protein? Would be interesting to do a protein tolerance test: drinking protein shake and then measuring insulin levels over time?

    • “If different people respond differently to carbs why would it not be the case about fat and protein?”

      I believe this is likely true; Gardner made the carbs/IR connection. Now need to figure out same for proteins and fats.

  • CynicalEng

    “The risk of hypoglycemia was similar in both treatments (48% vs 33% for FII vs carbohydrate counting, respectively, P=0.155).” and “significantly reduced the mean blood glucose level (5.7±0.2? vs 6.5±0.2?mmol/l, P=0.003)” so not a total fail.

    If insulin resistant people take in protein, do they get the extra blood sugar from the glucagon release and extra insulin, leading to an increase in BG ‘cos the insulin is ineffective ? Compared to normal / insulin sensitive people who show the same BG after protein intake despite insulin release (due to the counter effect of glucagon).

    • This has been explored in a variety of different conditions, including insulin resistance.

      Eg, see study by Gannon:

      After a protein bolus, blood glucose even tends to decline, despite increased glucagon.

      • CynicalEng

        Looks like I’ll be buying a box of test strips and re-running Gannon as n=1. My BG likes to draw horizontal lines I think my liver is over-enthusiastic.

    • Bob Hoskins

      Actually not a fail at all. “significantly reduced the mean blood glucose level (5.7±0.2? vs 6.5±0.2?mmol/l, P=0.003)” was followed by, “No severe hypoglycemia was reported.” in that study. Yes, they had an increase of readings under 70 mg/dL, but if none of them was severe, then this just represents closer to normal pancreatic blood sugar control.

      The variability in insulin requirements with various proteins has to to with the varying glucogenic properties of different amino acids.

      If I take BCAAs or glutamine as a supplement (pure, with no carbs) and don’t dose insulin, my blood sugar climbs just as fast as if I’d just eaten crackers. It isn’t just glucagon that is forcing the need for insulin.

  • Carole

    Bill, those that are holding for protein and having hypoglycemia are not doing it properly. I personally don’t recommend using an insulin pump to this end, but rather recommend Dr. Bernstein’s method: he uses Regular insulin and matches to a set meal of a fixed amount of protein with a fixed amount of non-starchy vegetables. His method is better more several reasons: 1.) It takes into account the dual nature of protein digestion which is first due to the Chinese food effect on T1 diabetics with glucose release from the bolus of food passing through the intestines and later about 3-5 hours out from the amount of glucose from digestion of the protein. 2. ) It also covers the digestion of the job starchy vegetables with its lower and slower action time. 3.) Possibly most important and misunderstood is that especially in long standing T1 and possibly small children who have sustained higher BG for a shorter duration, there is undiagnosed gastroparesis. This delay in digestion will cause hypoglycemia after a meal, especially when using a LC diet and trying to bolus for protein. The fix for this is to shift the shot of Regular insulin (normally taken 45 min before the start of the meal) up by 5 minute increments until it matches the curve of the meal and you have no lows. This may be after the meal. If this doesn’t work, there is a good chance that the gastroparesis is more severe and diet will require further modification to make it more digestible. The R-R study is used to properly assess GP and is little known or understood. This is why any study using standard methods of housing for protein do not yield usable or accurate results in my opinion.

  • Gerard Pinzone

    I tried digging through Good Calories, Bad Calories to see where Taubes specifically ties excess protein to weight gain, but I couldn’t find it. Do you have a page reference? I understand the assumption that the “carbohydrate hypothesis” implies insulin drives fat storage/weight gain and protein increases insulin, that Taubes must therefore imply that a high protein diet increases weight gain. A->B->C However, scanning for the word “protein” throughout GCBC shows examples of the reverse. (See pp 310-311.) I don’t think he ever says A->C, but in fact states the opposite.

    • “I’ve said there are some holes in Taubes’ theory, and while he never said to avoid protein-rich foods, this may be another one of those holes.”

      • I can’t see the hole though.

        As Gerard says – A->B->C may overlap on a simplistic Venn diagram, but this is hardly TWICHOO (™CarbSane).

        Point 8 of “Taubes’s Inescapable Conclusions” is closest to oversimplifying this:

        • perhaps I’m wrong, or maybe just didn’t phrase it correctly.

          In overfeeding studies: high protein = more muscle, less fat.


          • I don’t think that’s what either of us are talking about. Taubes may be off base on some stuff – he says so himself frequently – but I’m not sure which hole this protein/insulin [sans glucagon again :(] index thing is shining a light on.

          • lemme try this one more time:
            Taubes says that insulin is fattening. Protein induces insulin secretion, but protein isn’t fattening…

          • You know that, I know that, GP knows that… but when folk oversimplify the “insulin = fattening theory” as Taubes’s takeaway and therefore falsifiable cos protein it reminds me of Peter’s misgiving:

            “Whenever someone does a hatchet job on the carbohydrate hypothesis using the insulinogenic index of beef without mentioning glucagon I am left wondering why they were carrying an axe in the first place.”


            I’m no GT fanboi, but conflating the above A/B/C as a hole is basically the exact opposite of what he spent most of GCBC trying to get across.

            Not trying to nit-pick, it just jumped out at me (and GP it seems) as an Evelyn-ish statement. (*shudder*)

          • Hyperlipid should write Good Calories Bad Calories 2 with Dr. Lagokos doing the intro and doodle art 😉

          • I think Hyperlipid should turn his Proton series into a book!

          • Peter is explaining why protein-induced insulin secretion doesn’t cause hypoglycemia (because glucagon).

            In the Macleod study, all the meals contained carbohydrate; it’s possible FFAs weren’t suppressed after the higher protein meal because insulin levels were way lower:

            IOW, I don’t think glucagon has a [direct] effect on lipolysis (in humans), but haven’t seen too many studies on this.

          • melancholyaeon

            Agree. Just because folks not-so-bright misread GT doesn’t mean GT believes it. 😀 He’s updated his ideas a couple of times & readily admits when he’s wrong. For Crissakes he started an entire research thang to prove himself either right or wrong. And since he’s actually not currently testing an insulin-theories but is instead focusing on NAFLD, it looks like he’s given up on insulin; Attia doesn’t even eat low-carb any more. FWIW the one time I saw GT eat, he had an enormous piece of prime rib, a salad & a shot of tequila. 😉

          • CynicalEng

            Insulin from some carbohydrates is at way higher levels than from protein – beef steak insulin AUC at less than half that of white bread or potato in the Holt paper.

          • but theoretically, insulin’s impact on lipolysis is far more potent than on glucose uptake…


          • Thomas Hemming Larsen

            Come on. We all know that protein is like chocolate cake 😛

  • Hi.

    It’s interesting that white potatoes have a very high insulin AUC, yet they have a very high satiety index, as per

    I think it’s because the potatoes were “Peeled, boiled for 20 min, and stored at 4 °C overnight; reheated in a microwave oven for 2 min immediately before serving”.

    Potatoes produce a lot of resistant starch RS3 after overnight refrigeration. I reckon that this results in a sustained slow rate of glucose entry to the blood, after the initial spike from the non-resistant starch.

    What say you?

  • I’ve read/listened to everything I can find online from Rosedale and I’m 95% aligned with his stuff – but the mTor thing leaves me scratching my head as extrapolating too far with limited data.

    I fumbled the math a little (double the eggs) but I had this Q for Rosedale years ago:

    “For instance take an 80kg male who is maybe 60kg LBM, saying 1g/kg LBM means 60g of protein. For a single day or two I can imagine eating that little, but over time this is starvation unless you’re chugging a cup of oil per day.

    To put it in perspective – this is about 240-300g of beef, a fairly small steak. Ok that’s fine for breakfast, but then no more protein is allowed? Or to look at it another way, this is about 4-5 eggs. Can any 80kg bloke REALLY get by on just 4-5 eggs per day – that would require huge amounts of supplementary fat.”

    I’m all about getting as high fat as possible, but I just wouldn’t know what else to eat to get my 3-4k “calories” a day if I was limited to ~6oz of meat.

    Butter n booze, I guess.

    • This Old Housewife

      For a time, Jimmy Moore was doing this. He was Warrior dieting and eating 5 eggs/day along with a stick of butter–nothing else.

      • John Smith

        Ugh. That’s disgusting. I love eating keto but not THAT much.

    • “Or to look at it another way, this is about 4-5 eggs”

      dinosaur eggs?!

      • Haha yeah I said I messed up the egg math!

        Still, ~9-10 eggs is all the protein I’d be allowed – that leaves me with ~1,500-2,500 “calories” – so like 8-12oz of butter I guess?

        • “8-12 oz of butter”


          • Raymund Edwards

            Ron Rosedale in many interviews states He does not measure ketones and is more focused on restricting protein.

            However he allows many types of carbs. Hence in practice the carb intake is at the same level as the protein intake and might in many cases be higher.

            This allows a variety of food choices and nutrient sources and may provide a keto ratio of around 1 ( or less a bit like a LGIT diet for epilepsy )

            See macros here in his study
            Carbs = protein , FAT is not at high keto levels

            “” 20% carbohydrate, 20% protein, and 60% fat. “”


          • Edit: I’m going to math again as I fkd up some %ages

      • Yeah, circles.

        I want a practical example of what he’d consider a day’s worth of food if I’m consuming negligible carbs and am only allowed 50-60g of protein. I can’t picture it.

        • “I want a practical example of what he’d consider a day’s worth of food…”

          Even Jimmy eats more than 60g pro! (but you can probably get an idea by perusing his instagram account)

          • Yyyyyeah, not really interested in Jimmy fascination as a template. As I said from the start (years ago) – I want to know what Rosedale would prescribe for me, these ratios/”macronutrients” are all super vague – and completely untenable in a real world/real food situation without either supreme levels of fat supplementation – or hunger misery.

            I remember having a brief convo on facepalmbook with someone who works with him last year, he basically said “just wait for his next book which explains all this”. Guess I should look and see if it manifested.

          • Ah it was Jan this year. Dork on facepalmbook deleted the entire thread, but here’s the relevant bits from my email alerts:


          • Joseph E. O’Brien

            Not to pile on Jimmy – but from my vantage he suffers from sarcopenia – or at least appears to be significantly under muscled

          • RobJ

            I wouldn’t say sarcopecia altho like most ppl who don’t train/eat sufficient protein, he’s very slowly losing muscle mass as he ages.

            He’s just a dude who had below average muscle mass from the start. It was obvious when he got down to his lowest weight (220?) and was still easily 25%-30% body fat.

            I’m the same way altho I never got near his level of obesity. I’ve have been doing a controlled bulk in the mid-teens bf for the last few months and hypertrophy is exceptionally slow but w/e, better than regaining all that fat back.

            Jimmy’s attempt to gain muscle way back was huuuuge mistake. Just left him fatter as excepted. No one should try bulking above 20% bf and if even them, it needs to be a moderate and controlled surplus.

    • Tim Heineman

      I’ve had no trouble living with about 60-65 gms of protein per day, my LBM is 57 or 58 kg. But I don’t mess around with nuts the way Ron says. I use lard, suet, coconut oil, palm oil and butter. No olive oil. Never heard of the PUFs. 2 to 7% of my calories are carb. 80 to 85% oil. Roughly 2200 or 2300 Kcal satisfies my hunger for a day without trying to reduce quantity at all. I’m 66 years old. The nature of the fats, in my experience, plays a major role in satiety and therefore ad libitum Kcal consumed.

      I love cheese but it doesn’t satisfy me. Too much protein, I assume. But I’m not sure of the mechanism because, while the protein will draw insulin which will suppress blood glucose and so increase hunger, it will also draw glucagon so all bets are off.

      “Protein: 15% -> 30%: increased leptin sensitivity, reduced appetite, and weight loss (Weigle et al., 2005).” But Weigle et al. introduce their study with “Ad libitum, low-carbohydrate diets decrease caloric intake and cause weight loss. It is unclear whether these effects are due to the reduced carbohydrate content of such diets or to their associated increase in protein intake.” So they claim to be testing an effect of low-carb diets but the test is with a high-carb study! I’m not impressed.

  • Nagar J

    I recall Gary frequently stating that protein stimulates insulin release so much less dramatically than carbs that it’s not terribly important for metabolic health. Consistent with this, I also recall Gary describing his own diet as very meat heavy.

    To me it always seemed clear that Gary wasn’t blaming insulin per se for obesity and the diseases of civilization (as he is often misinterpreted) but rather chronic, diet-induced hormonal derangement, i.e. insulin resistance, metabolic syndrome. He obviously does not believe protein consumption plays a significant role in this process. He may or may not be correct in this hypothesis, but I don’t see a logical hole or inconsistency here.

  • Nagar J

    Bill, if you haven’t already, would be fascinating to get your take in more detail on the human evidence about protein —> mTor —> dying quicker. Rosedale is no dope, but I sure hope he’s wrong about this. 1 g/kg LBM protein + low carb = no fun at all to eat if the large jar of ghee isn’t handy at the moment.

    • yeah I definitely don’t think Rosedale is a dope by far. There is a lot in his protein article that is spot on.

      But I’ve yet to find a [human] intervention study showing a negative impact of higher vs. lower protein. I understand the theory, and maybe it’s just because it takes way too long to show accelerated aging in humans….

      Tl;dr: I think people can thrive on a wide range of protein intakes.

  • Kindke

    “”But condemning red meat and dairy because insulin is simply flawed.””

    Indeed I agree.

    I think too much attention is payed on the ratio of protein/fat when ppl should just be sticking with the rules of focusing on whole foods and keeping carbs/insulin low.

    As someone who is extremely sensitive to weight gain my experience is that meat and high protein foods are close to impossible to gain fat on yet if I add some tato’s or rice with my meat I quickly gain fat.

    by and large for me it seems to be something to do with glucose disposal, most dairy allows weight loss for me, but milk is *very* slightly fattening. Any food that requires high levels of glucose disposal quickly is immediately VERY fattening,

    consequently I notice that whenever im exercising heavily ( cycling to work/gym ) I can eat more carbs without fat gain, again supporting the idea that nutrient partitioning is whats driving my fat gain.

    theres lots of problems with insulin that goes beyoung simply inhibiting lipolysis and fat oxidation, it stimulates differentiation of pre-adipocytes and fasting hyperinsulinemia upregulates glut1 on adipocytes which isn’t going to be good for weight loss.

    I always come back to the FIRKO model, its adipocytes see zero insulin and consequently its adipocytes have virtually no GLUT1

    • “there are a lot of problems with insulin that go beyond simply inhibiting lipolysis and fat oxidation”

      replace “insulin” with “carbohydrate,” and that’s the picture I’m seeing, because I don’t see these things being problematic with protein-induced insulin secretion… highly suspect other factors/hormones at play.
      Eg, see protein overfeeding studies by Antonio, Bray, etc.

  • Pingback: Dieta low-carb e insulina - Primal Brasil()

  • John Smith

    I’ve lost a shitton of weight eating low-insulin foods and drinking low-insulin drinks, despite consuming above my TDEE, My TDEE is 2150 but I eat at least 3000kcal and the weight just falls off. And’s not simply water weight. Anybody who says low carb doesn’t have a metabolic advantage is either a liar or ignorant.

    • I don’t doubt the benefits of LC, but those numbers are preposterous!

  • Pingback: Lessons learned from my regain. – Fitter at Fortyish()