Published: April 25, 2025
The hormone insulin is key for regulating your blood glucose levels. You may have heard that a reduction in insulin sensitivity, commonly referred to as insulin resistance, plays a major role in the development of prediabetes and type 2 diabetes. But that is only one part of the story. Often overlooked but similarly important is the other part – your body’s ability to produce insulin.
Because it’s the balance between your insulin sensitivity and your beta-cell function — your ability to produce insulin — that determines whether your blood glucose stays in a healthy range or starts to climb, which is what we see in prediabetes and type 2 diabetes.
So, in this blog post, I discuss a simple blood test that can reveal both your insulin sensitivity and your beta-cell function. And I also explain why understanding BOTH your insulin sensitivity and your beta-cell function is essential for protecting your long-term health.
What is Beta-Cell Function, and Why Does It Matter?
Beta-cell function is a measure of how much insulin your pancreatic beta cells produce relative to how insulin-sensitive you are. And the amount of insulin produced by your pancreatic beta-cells in response to a meal is directly related to the degree of insulin sensitivity. In healthy people, the more insulin sensitivity declines, the more insulin the beta-cells need to procuce to compensate.
This is a bit of a complex topic, so let’s draw up a simple graph, without units of measurement, but just to illustrate the concept. Let’s assume you are very insulin sensitive. In this case, your beta-cells will produce very little insulin after a meal, even a meal rich in carbs. That is not because they cannot produce more, but because they don’t HAVE to produce more. As long as you have enough insulin in your blood relative to how sensitive your tissues are to that insulin, your body will be able to clear glucose quickly after a meal, and you will maintain normal glucose tolerance. So this means that if your insulin sensitivity is high, your insulin production can be low simply because your cells don’t need a lot of insulin to respond to it.
To find an analogy here, if you can hear very well, you can hear music even if it’s very quiet. And only if your sense of hearing declines do you need to turn up the volume. And that works, up to a point, for you to still be able to listen to music even as your sense of hearing declines a lot. It’s really very similar with insulin sensitivity and insulin production.
In line with that, let’s assume you gain a bunch of weight, and your insulin sensitivity decreases. In other words, you become moderately insulin resistant. Now your tissues will be less sensitive to insulin, and each molecule of insulin will be less effective at shuttling glucose from the blood into the cells. That means you need to secrete more insulin to compensate for this. So the question now is whether your beta-cells can produce more insulin to fully compensate for this. If they can, you will remain glucose-tolerant. That is illustrated by the green dot in the figure below. If they can’t, you will become glucose-intolerant to some degree, and your blood glucose levels will become elevated. That is illustrated by the orange dot in the figure below.
In essence, you could become quite insulin-resistant, which is not great, but as long as your pancreatic beta-cells can fully compensate for this insulin resistance by producing more and more insulin, you will remain fully glucose tolerant and your blood glucose levels will remain in the normal range. So, no matter where in the lightly green-shaded area in the figure below your insulin sensitivity and beta-cell insulin production fall, be it very insulin sensitive or very insulin resistant, as long as your body makes enough insulin for your level of insulin sensitivity, you will remain glucose-tolerant.
Now, obviously, in this figure, a person with low insulin sensitivity will produce much more insulin after a meal than a person with very high insulin sensitivity. However, they all have one thing in common: they all make enough insulin for their level of insulin sensitivity. We could say that their beta-cell function is 100% of what it needs to be, considering how insulin sensitive they are. Keep this in mind, because it is going to become relevant when we discuss how you can measure your beta-cell function.
On the other hand, if your insulin sensitivity decreases for any reason and your beta-cells can NOT fully compensate, as is illustrated by the orange dots in the figure below, that’s when you have too little insulin in your blood relative to how insulin sensitive your tissues are. In those cases, you would become prediabetic, as illustrated by the orange-shaded area. What all of these orange dots have in common is that the amount of insulin produced by the beta-cells is insufficient given the specific level of insulin sensitivity. So beta-cell function may be just 90%, 80%, 70%, or 60% of what is needed here for everyone within that orange-shaded area, and that’s why these people would have moderately elevated glucose levels, as in prediabetes.
The red dots in the figure below now show the more extreme version of this defect: now, insulin production is way too low for any given level of insulin sensitivity, and all of these people represented by these red dots have diabetes. What they have in common is that their beta-cell function is only 50% or less of what is needed, given their level of insulin sensitivity. The further they are in the lower left corner, the worse we would expect their glucose tolerance and glycemic control to be.
I hope this illustrates that beta-cell function, i.e., the ability of your beta-cells to produce as much insulin as needed, is just as important for blood sugar regulation as insulin sensitivity. And that beta-cell function always needs to be considered relative to the degree of your individual insulin sensitivity.
If this is not fully clear yet, I recommend my blog post about the Regulation of Blood Sugar in which I explain all of this in much more detail.
OK, so if beta-cell function is so important, how can you figure out whether your body is currently able to produce enough insulin for your level of insulin sensitivity? That’s what we are covering next.
How to Measure Beta-Cell Function and Insulin Sensitivity
In a previous blog post, I shared how to measure insulin resistance using fasting glucose and fasting insulin concentrations. This measure of insulin resistance is called the homeostasis model assessment of insulin resistance (HOMA or HOMA-IR).
There is another way to use fasting glucose and insulin to measure your insulin sensitivity and also obtain a measurement of your beta-cell function. This measurement is based on a more advanced HOMA-model, sometimes called HOMA2 or the HOMA computer model. That is because it relies on a computer-based algorithm to calculate HOMA-%S, which is a measure of insulin sensitivity, and HOMA-%B, a measure of beta-cell function. Both are expressed in % of normal, with 100% set to normal.
Now, to use the model, you need to purchase a license, which most doctors’ offices do not have access to, which greatly limits its use. That is highly regrettable, because we can learn a lot from these data. One workaround is the figure below, which is based on the prior HOMA1 model (sorry, this was stated incorrectly in the video; unfortunately, I cannot share anything about the HOMA2 model because it is a licensed product). You can download a high-resolution copy of this figure for your own use here.
So, how does this work?
Well, first, you need to have glucose and insulin concentrations in fasting blood measured. Ideally, do this in the morning, after an overnight fast of at least 10 hours. Ideally, you’ll have your blood drawn in the morning of a normal day. Normal day here means that you have not eaten an unusually large or late dinner the night before, have not had an unusual amount of alcoholic beverages the night before, have NOT been sick for a few days, have not had surgery, and have not been unusually active or inactive in the few days before the blood draw. In other words, you want to measure this on a day that is like most of your days.
OK, now you have a fasting glucose and a fasting insulin measurement. If you have a fasting glucose reading in mg/dL, divide that number by 18 to convert it to mmol/L.
Then, print the figure and add a dot where your fasting glucose level, in mmol/L, and your fasting insulin level intersect. I have done this in the figure below for Mike. He has fasting glucose of 4.4 mmol/L, which is 79 mg/dL, and fasting insulin of about 5 uU/mL. So you see that red dot shows that Mike is fully insulin sensitive, meaning his insulin sensitivity, or HOMA-%S, is 100% of normal. See the dashed red line. He also has 100% beta-cell function, indicated by his HOMA-%B, which is 100%. See the purple line.
Now, the important part: his beta-cell function of 100% does NOT tell us anything about how much insulin Mike’s beta-cells can maximally produce. HOMA-%B of 100% simply means that at his current level of insulin sensitivity, his body can produce enough insulin to keep blood glucose in the normal range. And yes, Mike’s glucose tolerance and glycemic control are very good by all indications, such as his fasting glucose of 4.4 mmol/L or 79 mg/dL, and his HbA1c of 4.8%.
By the way, in case you are wondering how HOMA-%S and HOMA-IR are related, HOMA-%S is simply 100 divided by HOMA-IR. If you recall, HOMA-IR of 1 is normal, and HOMA-%S in that case, would be 100 divided by 1, which is 100%. HOMA-IR of 2 is equal to HOMA-%S of 50%, because 100 divided by 2 is 50. So, in a way, HOMA-IR and HOMA-%S tell us the same thing, just expressed differently.
OK, so Mike has normal everything: normal insulin sensitivity, normal beta-cell function, and – as a result of that – normal glucose tolerance and normal glycemic control.
One quick disclaimer here: as I have shared previously, the gold-standard tests for insulin sensitivity and beta-cell function are quite involved and are rarely performed in clinical care. HOMA-%S and HOMA-%B are correlated with the gold-standard measures, but they do have their limitations. Therefore, it may be more adequate to think of these as estimates of your insulin sensitivity and beta-cell function. Don’t place too much emphasis on minor deviations from the norm. These could just be measurement errors. Instead, use this figure to get a sense of where you roughly lie, so that you can understand whether you may have insulin resistance, beta-cell dysfunction, and glucose intolerance, or not.
Importantly, in the area of the graph where Mike is located, around HOMA-%S and HOMA-%B of 100%, you will note that small differences in fasting insulin and fasting glucose can lead to substantial changes in HOMA-%S or HOMA-%B. This is the major limitation of this method because both fasting insulin and glucose can vary quite a bit even from one minute to the next. For example, if you feel anxious or stressed, or if you need to rush to get to the lab for the blood draw, or if you had a very large meal the night before or a lot of alcohol, all of these could affect your fasting glucose and insulin concentrations. You can minimize these variations by making sure to follow the instructions about the blood draw above. Further, you can interpret HOMA-%S and HOMA-%B in the context of other measures of glucose tolerance, such as HbA1c or any CGM data you may have. For example, if your CGM data show that your glucose levels are in the 70-140 mg/dL (3.9-7.8 mmol/L) range 98%+ of the time, you never have any glucose spikes to 180 mg/dL (10 mmol/L) or higher, and your glucose levels are consistently back to your baseline two hours after starting a meal, this is strongly indicative of perfect glucose tolerance, as explained in a prior blog post. So in this scenario, you would expect HOMA-%B to be at least 80% of what it needs to be; if it came out lower than that in this test, this could be due to random variation in your fasting glucose and insulin levels.
How Shifts in Insulin Sensitivity and Beta-Cell Function Affect Glucose Tolerance
OK, next, let’s take a look at Jake. Jake is a member of the Nourished by Science Prevention Program, where we help members make diet and lifestyle changes to reduce their risk of chronic disease. Jake is a real person, even though I’ve changed his name here for privacy reasons. When Jake joined our program in July of last year, his fasting glucose was 12.8 mmol/L, which is 231 mg/dL, and his fasting insulin was 21.8 uU/mL. His HbA1c was around 10%, so this means he had poorly controlled type 2 diabetes. And that was even though he was on several anti-diabetes medications.
At that time, Jake’s HOMA-%S was at about 10% of normal, which means it was reduced by 90% from normal. And at that level of insulin sensitivity, his beta-cell function was around 40% of normal, or maybe 42% of normal. We would interpret this to suggest that Jake was substantially insulin-resistant, and that at this low level of insulin sensitivity, his beta-cells were not able to make nearly enough insulin to keep his blood glucose controlled within the normal range. And that is exactly why he was very glucose intolerant, and why his blood glucose levels were chronically elevated.
Now, importantly, Jake may actually have had higher insulin levels after his meals than Mike from our earlier example (see figure below). This is something that is often misunderstood when people talk about beta-cell function, so let’s clarify this. Mike’s HOMA-%B, so his measure of beta-cell function, was 100%, whereas Jake’s HOMA-%B was at around 42%. Some people take this to indicate that Jake’s insulin levels after a meal were 42% relative to Mike’s, in other words, 58% lower. That is NOT the correct interpretation here. Instead, what we can say is that Mike’s beta-cells are able to produce 100% of the insulin that is required for his level of insulin sensitivity. Because he is very insulin sensitive, he does not need to make a lot of insulin, so a HOMA-%B of 100% in Mike’s case may still mean fairly low insulin levels after each meal. In Jake’s case, he was able to make 42% of the amount of insulin needed for his level of insulin sensitivity. However, the important point to understand is that Jake was much less insulin-sensitive than Mike, and therefore needed much more insulin, actually many times more. And even though he may make more insulin, it’s not nearly enough for his level of insulin sensitivity to keep his blood glucose regulated within the normal range. So, keep in mind that this measure of beta-cell function, HOMA-%B, always needs to be interpreted relative to the degree of insulin sensitivity.
OK, now, back to Jake. As a member of the Nourished by Science Prevention Program, he made numerous changes to his diet and lifestyle to address the root causes of his insulin resistance. And as he made these changes, Jake’s insulin sensitivity gradually improved. Eight months later, he had improved quite a bit; see the figure below. His fasting glucose was now around 5.8 mmol/L, or 105 mg/dL, and his fasting insulin was now around 13 uU/mL. His HOMA-%S, as an estimate of his insulin sensitivity has improved to about 30% of normal, from 10% of normal. And even though he still was fairly insulin resistant, HOMA-%B, the estimate of his beta-cell function, was now close to 100%.
How we interpret this is that at his current level of insulin sensitivity, Jake’s pancreatic beta-cells are able to make enough insulin to keep his blood glucose levels mostly regulated within the normal range. As a result, his HbA1c has improved to 5.8%. And that is even though he had to lower his anti-diabetes medications several times. He is now just so outside of the normal range, so still slightly glucose intolerant, but his glycemic control is much, much better than before. In fact, remember that HbA1c provides an estimate of the average blood glucose concentration over the past 3-4 months, and it seems to me that Jake’s glycemic control continued to improve quite a bit during this period, so his HbA1c of 5.8% probably doesn’t even fully reflect his actual improvement in glycemic control at this last time point. Independent of that, however, it’s clear that Jake is on a great path, and if he improves his insulin sensitivity at least a little bit more, he will probably be able to fully normalize his glucose tolerance and glycemic control.
Okay, next, let’s take a look at another member of the Nourished by Science Prevention Program. Let’s call her Jane for privacy’s sake (figure below). Jane’s fasting glucose is 5.4 mmol/L, which is equivalent to 97 mg/dL. That is in the normal range, and if that’s the only measurement done here, Jane or her doctor wouldn’t know that she is metabolically not very healthy. Take a look at her fasting insulin levels, which are around 37 uU/mL. Jane is massively insulin resistant, with HOMA-%S at maybe around 9% of normal, similar to where Jake initially was.
Take a look at the red line that I have drawn in the figure above to show HOMA-%S of 10%. Jane is actually a little bit less insulin-sensitive than Jake. However, and that’s the big difference between Jane and Jake, Jane has very, very healthy beta-cells, and so she can fully compensate for her degree of insulin resistance. Jane’s HOMA-%B seems to be at around 250% of normal, so her beta-cells can make huge amounts of insulin. If she eats a meal rich in carbs, her blood insulin levels will be very, very high. High enough, in fact, to keep her blood glucose levels more or less in the normal range all of the time. Remember that Jake’s HbA1c at that time was around 10%. Jane’s is 5.8%, just so outside the normal range, and she probably has prediabetes. Very often, even if this HbA1c level is measured, a doctor will not do anything about it, and if fasting insulin is never measured, no one will ever know what is actually happening with her insulin levels after every meal. As I will explain later in the blog post, it is my opinion that it is as important for Jane to improve her insulin sensitivity as it is for Jake. Being so insulin resistant and having chronically elevated insulin levels is far from ideal for our long-term health.
But before we discuss that, let’s take a look at another example: Zoe. She is, in a way, the opposite of Jane. Zoe’s fasting glucose is 5.2 mmol/L, or 94 mg/dL. Her HbA1c is 6.0%.
Her elevated HbA1c suggests that she has prediabetes. Now, interestingly, Zoe’s fasting insulin is at around 4 uU/mL, which is very low. What does that mean? Well, it means that her insulin sensitivity, HOMA-%S, is at around 100% of normal, so this is not the primary cause of her glucose intolerance. However, her HOMA-%B is around 50% of normal, meaning that at her very good level of insulin sensitivity, she can not make nearly enough insulin to keep her blood glucose in the normal range. We say that Zoe has beta-cell dysfunction, and that is why she has prediabetes. Now, Zoe would need to approach her glucose intolerance quite differently than Jake, and we’ll talk in a future blog post about causes of beta-cell dysfunction, and what can be done to improve the ability of our beta-cells to secrete insulin.
For now, the best thing Zoe can do to keep her blood glucose regulated in the normal range would be to lower the glycemic load of her diet, i.e., to eat fewer carbs and particularly to stay away from carbs with a high glycemic index. For a patient like that, I would also suggest always starting a meal with a serving of fiber-rich non-starchy vegetables or a salad with vinaigrette, and also with a serving of a protein-rich food. That prepares the body for any carbs that may be consumed later in the meal and minimizes their blood sugar impact.
How to Use Continuous Glucose Monitor Data to Detect Beta-Cell Dysfunction
There is another way to detect poor beta-cell function, and that is based on data from a continuous glucose monitor, or CGM. In people who have a diminished ability to secrete insulin, particularly in the first-phase, we tend to see the pattern below when someone eats a meal rich in carbs: glucose levels shoot up like a rocket to a level of 180 mg/dL, 10 mmol/L, or higher. Often, this occurs within the first 30 or 40 minutes after starting a meal.
If you see the pattern shown in the figure above, i.e. a rapid rise to 180 mg/dL or 10 mmol/L regularly, that could be indicative of a diminished first-phase insulin response. Again, I’ll talk about potential causes of beta-cell dysfunction and particularly a diminished first-phase insulin response, and what can be done about it, in a later blog post.
Why Maintaining Normal Insulin Sensitivity and Beta-Cell Function Is Critical for Long-Term Health
Okay, so in the case of Jake or even Zoe, I think most people will understand that it’s important for them to keep their blood glucose levels consistently in the normal range. That is simply because chronically elevated blood glucose levels, as in prediabetes in Zoe’s case or type 2 diabetes, as in Jake’s case, are a major risk factor for numerous chronic diseases.
However, I argue that it is almost as important for Jane to improve her insulin sensitivity as it is for Jake, and I would say that even if Jane’s HbA1c level were totally in the normal range. Because Jane’s risk of chronic disease is substantially elevated because she is so insulin resistant. We understand this best if we look at WHY someone like Jake has an increased risk of chronic diseases.
In Jake’s case, his long-term risk of suffering from another chronic disease will be clearly elevated because he has type 2 diabetes. When we talk about chronic disease here, we mean that people with type 2 diabetes clearly have an increased risk of cardiovascular diseases, such as a heart attack or a stroke, as well as chronic kidney disease, liver disease, and several types of cancer. And so, this is clearly important to understand: if someone has any form of diabetes, including type 2 diabetes, they do have an elevated risk of several chronic diseases because of their chronically elevated blood glucose levels. However, and this is very important to understand, that’s only part of the story.
So, yes, if blood glucose levels are chronically elevated, we should aim to bring them down, ideally into the normal range. However, as I hope has become clear in this blog post, when we are glucose intolerant and have chronically elevated blood glucose levels, there is a reason for that: we cannot produce enough insulin for our level of insulin sensitivity. In type 2 diabetes, this usually means that we suffer from some degree of insulin resistance plus some degree of beta-cell dysfunction. I argue that we should pay as much attention to insulin resistance as we do to chronically elevated blood glucose levels.
For one, insulin resistance itself is a risk factor for many chronic diseases. Further, if we are insulin-resistant and our beta-cells can at least partly compensate for this by producing more insulin, that means we may have chronically elevated insulin levels. We call this hyperinsulinemia. And such chronically elevated insulin levels are also a risk factor for many chronic diseases.
But let’s also consider WHY someone is insulin resistant. As I have discussed in prior blog posts, there are many potantial causes of insulin resistance, including chronic inflammation, excess visceral fat, excess ectopic fat, such as in the liver or musculature, low muscle mass, a sedentary lifestyle, different micronutrient deficiencies, certain medical conditions, such as sleep apnea, or sleep deprivation. And without exception, ALL of these in one way or another are ALSO risk factors for various chronic diseases. So, if someone has type 2 diabetes because they are insulin resistant because of one or several of these reasons, if we want to minimize their risk of chronic disease, we must normalize their blood glucose levels. But we also should aim to reverse their insulin resistance and address the root causes underlying that insulin resistance.
And in extension of that, I argue that even if someone is insulin resistant, but their beta-cells can fully compensate by producing more insulin and as a result, they do NOT have diabetes or prediabetes, i.e., their blood glucose levels remain normal, they likely still have an elevated risk of chronic diseases because they must be exposed to one or several of these root causes of insulin resistance here. And they certainly are exposed to insulin resistance and the resulting hyperinsulinemia.
Summary & Conclusion
We discussed that you can use your fasting glucose and insulin measurements to get an estimate of your level of insulin sensitivity and your level of beta-cell function.
The measure of insulin sensitivity is called HOMA-%S, and is relative to normal insulin sensitivity, which is defined as a HOMA-%S value of 100%.
The measure of beta-cell function is HOMA-%B, and again, normal means 100%. In this case, HOMA-%B needs to be interpreted relative to our specific level of insulin sensitivity. A HOMA-%B of 100% or more means that our pancreatic beta-cells can make enough insulin for our level of insulin sensitivity. A HOMA-%B lower than 100% means that our pancreatic beta-cells can NOT make enough insulin for our level of insulin sensitivity. In those situations, we are usually glucose intolerant to some degree.
If you want to download this poster, you can do so using this link HERE.
Also, if you’re interested in joining the Nourished by Science Prevention Program and working with me directly to minimize your chronic disease risk factors, click the link HERE to add you name to the waitlist. We open the program every two months for new enrollment.
Understanding whether we have reduced insulin sensitivity and/or reduced beta-cell function is a first step to getting to the root cause or causes of any blood sugar issues we may have. If you have reduced insulin sensitivity, I recommend this blog post next on how to reverse insulin resistance, and if you have reduced beta-cell function, you may benefit from applying the strategies I share in this blog post about how to avoid blood sugar spikes.
With that, let me thank the Patrons who support the generation of free content on this website and in the Nourished by Science YouTube channel. If you value our content, please consider becoming a Patron of the channel.
References
- Kahn et al.; Quantification of the relationship between insulin sensitivity and beta-cell function in human subjects. Evidence for a hyperbolic function. Diabetes 1993; 42: 1663-72.
- Matthews et al.; Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28: 412-9.
- Park et al.; Assessment of insulin secretion and insulin resistance in humans. Diabetes Metabolism Journal 2021; 45: 641-54.
35 Responses
Hi Mario. Thanks as always for a very informative post. I wanted to ask about your CGM graph:
Your post discusses the height for the peak and the pace of increase. I wanted to ask something about shape as well: I noticed in my CGM readings a double-humped peak: glucose goes up (say to e.g. 7.5), then very rapidly down to e.g. 6.5, and then climbs again – sometimes almost to the same height (7.5 in this example), and only then starts to go down to baseline. I read somwhere that this pattern is indicative of effective first-phase response. Is that the case? What else can such pattern tell us? Thanks!
Hey,
Good question. We call this pattern a bi-phasic glucose response, and it can indeed be a sign of excellent glucose tolerance due to robust first- and second-phase insulin responses. However, this should be interpreted only as a bi-phasic glucose response if you do not exercise after the meal. Because if you went on a short walk after a meal, that could bring your glucose down due to the insulin-independent uptake of glucose into contracting muscle cells. And if you then arrive back home and sit down and your glucose absorption from the GI-tract is still not completed, that could cause the second glucose hump. If you do not exercise and routinely see a bi-phasic glucose response, then yes, congrats for solid beta-cell function in both the first and second phase.
Cheers
Mario
Thank you Mario! This was the video I have been waiting for ! After finding out my HbA1c rose to 5.8 but my fasting glucose was 80 and knowing that I didn’t have any of the factors that cause insulin resistance, I got a CGM. Using your previous videos, I tried to analyze my curve shapes and was able to figure out that I had a loss of beta-cell function ( I have a history of gestational diabetes). This was the missing video in my educational journey. I had already been eating a diet that was focused on blood sugar regulation which keeps me mostly in the normal range, but realized that I need to be a bit stricter and avoid the occasional dessert or white flour foods as well as exercise or move after each meal. I am so looking forward to your upcoming videos about this topic.
Hi Dee,
I guess we all like to have something sweet occasionally, so what you could experiment with is to have large amounts of fiber-rich vegetables and also protein-rich foods first in the meal, and then add any carbs (dessert, or a small piece of dark chocolate) later in the meal. The fiber and protein will delay the digestion and absorption of the carbs, and the amino acids in protein will also trigger the release of insulin by the beta-cells. That insulin then helps handle the incoming glucose from the carbs consumed later.
Cheers
Mario
Thank you Mario. That is so interesting. I love the examples and I think I might be like Zoe. I first noticed my fasting glucose was a little high a couple of years ago – 111. My doctor did not mention this but I thought it might be a good idea to work on it. This led me to try finger pricking and then later a CGM a few times. Using Levels Health I was able to have my glucose and insulin tested and HBA1C was 5.7 and insulin was 5.5. My glucose does tend to go pretty high pretty quickly after carbs but it didn’t really seem like I had insulin resistance. I tried to eat fewer carbs and then I lost about 15 pounds without really meaning to. I didn’t really want to lose more and I missed carbs – so I try harder to follow your suggestions to lesson spikes. This has mostly helped a lot but I am still curious of your future post regarding beta-cell dysfunction.
Hi,
Thank you for your feedback. If you didn’t want to lose weight, and maybe didn’t even have much weight to lose, then this could have even further contributed to glucose intolerance because any weight loss is always associated with a loss of muscle mass. And muscle is the tissue that removes most of the glucose after a meal.
Also, a low carb-diet is likely to reduce beta-cell function further. See my comprehensive response to Axel in the comments here where I explain this in much detail. Now, that said, a lower glycemic load diet may still be a good choice for people with poor beta-cell function; however, make sure to keep the carb content low at all meals if you do that. If you want to eat carbs, then try to eat similar amounts of mostly low-glycemic index carb foods with each meal.
Cheers
Mario
That is something I had not thought of – losing weight equal losing muscle and that would relate to glucose intolerance. Wow! But that might also explain why older people who tend to lose muscle, end up with more issues. I am 71 . . . went from 132 pounds to 119 – so BMI from 22.7 to 20.4. At that point I decided I just needed to eat a lot more plus carbs, along with following your suggestions in the “avoiding spikes” blog. I read your response to Axel and also remembered your pointing out the second meal effect. My weight is holding around 120 and at least I lost the weight around my middle section! I am primarily a hiker but am doing some light weight lifting – maybe that will help but I have never been very muscular and my joints can’t seem to handle lifting heavy weights. Another thing I recently became aware of is my vitamin B12 is low at around 196 pg/ml. I googled that and it appears that B12 can have an effect on glucose . . . I have started supplementing B12 so maybe that will help. I really appreciate your blogs and all the detailed information. Plus the questions by others and your answers. Thank you, Cathy
Hi Cathy,
I have come to think of weight loss as a double-edged sword. Yes, it can be very beneficial to many people to reduce any visceral or ectopic fat they have. However, that weight loss would ideally be paired with efforts to maintain as much muscle mass as possible by eating enough protein and engaging in regular resistance training. If you are uncertain about lifting weights, you can do simple body weight exercises. As long as you do them often enough (3 times per week would be great) and with enough intensity (the muscles should get tired during the exercise), this can have a huge impact on our quality of life and our health as we age. Consider also that how strong our muscles are plays a major role in determining how strong our bones are (because the bones make themselves stronger to resist the pull by stronger muscles). Here are a few YouTube channels that seem to provide solid instruction for older people who want to exercise more regularly:
Senior Shape Fitness:
https://www.youtube.com/@SeniorShapeFitness
HASfit:
https://www.youtube.com/@HASfit
Improved Health:
https://www.youtube.com/@ImprovedHealth
HT Physio – Over-Fifties Specialist Physio:
https://www.youtube.com/channel/UCXGeAZFwWu-zMxskaDuXjmg
Cheers
Mario
Conneal’s post is my story except I don’t know the insulin number yet. Waiting for naturopath appt to get a test requisition since doc didn’t think it was important – “Just reduce your carbs” – and here I am 10lbs lighter :/ I guess I will have to add weight training to add muscle mass back in addition to your other suggestions. Thank you so much and God bless!!
Much thanks for the video links below! I really appreciate it and look forward to reviewing and adding some routines that might help both glucose control and my bone density issues.
Really good indeed! I’m a T2D on a strict diet and recently started using sensors which certainly opened a window to how my metabolism works. I’ve done a number of personal trials with different types of nutrients reaching standard results with high glycemic
-index foods helping to much reduce glucose postprandial peaks. Nevertheless, I’ve observed an interesting phenomenon with alcohol as it reduces rapidly my glucose to a normal level around 5.5 mmol/L from 12mmolL. My standard plasma glucose response to meals rich in carbs is consistent with the one you point as insufficient beta-cell function. However, this effect of small amounts of alcohol on down-regulating glucose levels seems at odds with an insufficient insulin release as the glucose descent is in a mirror image of a poor beta cell function. There is some literature over the effects of alcohol on plasma glucose but no much on its mechanism of action. My hunch is that alcohol increases the sensibility of beta cells prompting the release of insulin rather than improving insulin resistance. In that case, beta cell dysfunction wouldn’t be primary and therefore treatable(not by prescribing alcohol which has too many undesirable side effects). I’d be extremely grateful on your comments Mario!
Hi Enric,
Thank you for the comment. I will be honest, I have not reviewed the impact of alcohol on glucose tolerance in a long time, and am hesitant to say something about it before educating myself. I have added this as an idea for a future video to my list, and will address it in due time. It seems to me that this should be of interest to many people.
Cheers
Mario
Why use complicated graphs to deduct HOMA-%S and HOMA-%B and not just check a C-peptide to measure beta cell function?
Hi Andrei,
If you know how to interpret C-peptide relative to your insulin sensitivity, then C-peptide is also a good option, possibly even better, for assessing your beta-cell function. The problem is that there is no good measure of insulin sensitivity that is based on c-peptide, and people would therefore also have to measure fasting insulin and glucose, and I dare say it’s going to be difficult to convince most doctors to measure glucose, insulin, and c-peptide. I therefore feel that calculating HOMA-%S and HOMA-%B is a reasonable compromise that is more feasible for most people while also being fairly easy to interpret.
Cheers
Mario
Mario,
I actually just received my C-peptide and fasting insulin + glucose levels. Are there any resources you recommend for interpreting the C-peptide results in the context of my insulin and glucose? Unfortunately, my C-peptide is very low, 1.0 ng/mL.
Thank you for putting together this wonderful blog post and video. Based on the chart (insulin 4.0 uIU/mL, glucose 4.3 mmol/L), it seems like beta cell dysfunction is primarily driving my pre-diabetes (A1C 5.8). This is also consistent with previous CGM readings, which revealed spikes in excess of 200 mg/dL after carb-heavy meals, but otherwise normal-looking levels. Looks like I really need to stay on top of managing my carb intake and incorporating fiber and protein into my diet.
I very much look forward to the future post/video on the causes behind beta cell dysfunction.
Ken
Ken,
Sounds to me like your C-peptide is confirming what HOMA-%B is suggesting, right? You can use C-peptide as well to assess beta-cell function, and – in fact – it is preferred for this purpose over insulin. However, we don’t have an easily interpretable way to relate C-peptide to a measure of insulin sensitivity without also measuring insulin.
Best,
Mario
Good video as usual. The fact that you don’t mention it means that there is little one can do to improve beta cell production? Avoiding spikes only treats the symptoms not the cause, right? My CGM data look a lot like what you show. Guess i should have those Homs Tests done next time at the doc
Hi Axel,
There are a few things that cause beta-cell dysfunction that are reversible, and I’ll cover these in a separate video. However, beta-cell dysfunction has a stronger genetic component than insulin resistance, and often it’s much easier to improve glucose tolerance by aiming to improve insulin sensitivity than insulin secretion.
One important thing that is often under appreciated isa that a low-carb diet reduces the amount of insulin that can be secreted in the first-phase insulin response. Think about it this way:
– When we eat carbs, the glucose from these carbs can sometimes enter our blood stream within a few minutes. Think, for example, about a sugar-sweetened beverage rich in sugar. These beverages do not stay in the stomach, and the sugar doesn’t require much digestion, so it can be absorbed very rapidly.
– Insulin is a protein, and our pancreatic beta-cells will start the process of making the insulin protein when we ingest carbs. That takes quite a few minutes, because the part of the DNA that encodes the insulin protein needs to be read and transcribed to mRNA, and then the mRNA needs to be translated into protein. And then than protein needs to be secreted into the blood. If this was the only mechanisms through which insulin was secreted when we eat carbs, then our blood glucose level would regularly exceed 1,000 mg/dL (one thousand, yes) after a meal rich in carbs.
– Therefore, the body had to come up with a way to secrete insulin into the blood much more quickly whenever we eat carbs. That mechanism is that there is always some small amount of insulin stored inside of the beta-cells, and that insulin can be secreted into the blood stream within seconds of consuming carbs. That is what we call the first-phase insulin response. The second-phase insulin response then is the insulin that needs to be made from scratch. Think about this a bit like a factory with a warehouse. The insulin made in the first phase is insulin that is sitting in the warehouse that can be shipped immediately when the order comes in. The insulin that can be made in the factory takes a while to be assembled.
– So if you regularly eat carbs in every meal, the body learns that carbs are regularly coming in, and stores a lot of pre-formed insulin in the warehouse. Therefore, the first-phase insulin response is robust. However, if you are on a low-carb diet, the body stops keeping insulin in the warehouse, and therefore the first-phase insulin response is diminished. That means that if a person on a low-carb diet unepectedly eats a sandwich or rice or pizza, they will have too little insulin in blood in the first few minutes after eating, and they could develop a blood glucose spike.
– This effect of a low-carb diet on the first-phase insulin response is reversible, but it can take 2-3 weeks to recover a normal first-phase insulin response. It would therefore be important to re-introduce carbs slowly after a low-carb diet has been followed for a while.
Cheers
Mario
That’s a very helpful explanation for the reasoning behind including carbs in every meal.
Has there been any research as to whether this strategy also works when there is a congenital defect in the beta cells, e.g. some forms of monogenic diabetes? Based on family history and personal clinical data, my doctor suspects MODY3 (HNF1a mutation) has been in the family for at least 3 generations, although we’re still waiting on confirmation. At this point, I still have some endogenous insulin production, but it does appear to be significantly limited despite not having any signs of insulin resistance.
Hi Ori,
To make sure I am not misunderstood: low-carb or even very-low carb is a good option, and in many cases even best for people with glucose intolerance. The point I wanted to make is that if someone is generally consuming a low-carb diet, they should make sure to keep every meal low-carb and not constantly switch between low-carb and high-carb meals. The reason is that in that case, the first-phase insulin response would be diminished in response to the low-carb meals (because the prior meals were low in carbs), which would then potentially cause a blood glucose spike.
If someone has reduced beta-cell function for reasons such as autoimmune destruction of the beta-cells or genetic factors (such as MODY, but also other, less clearly defined genetic variants), they would NOT be able to increase their insulin secretion beyond a certain point just by eating carbs. Look at it this way: every person has what we call a maximal beta-cell insulin secretory capacity (we can even measure this in clinical research by maximally stimulating the beta-cells). That maximal output is your baseline. In some people, this is very high. In others, it’s genetically low or very low. And again in others, it is lowered over time by autoimmune processes, and also by processes such as fat accumulation in the pancreas (which is probably reversible in many people). By consuming carbs with each meal, you ‘motivate’ your beta-cells to keep insulin secretion at that high level. By contrast, if you now go on a low-carb diet, then the pancreas learns that no carbs are consumed, and lowers the amount of pre-formed insulin kept in the beta-cell that can be secreted as part of the first-phase insulin response. This is not a problem, as long as all meals are low in carbs. And this phenomenon is also reversible by gradually increasing carb intake.
Now, what does this mean for someone like you who may have a genetically limited maximal insulin secretory capacity?
1.) It definitely means that you benefit from maintaining normal insulin sensitivity to reduce the amount of insulin that your body needs to maintain glucose homeostasis.
2.) You probably also want to minimize consumption of ‘naked’ carb meals, i.e., meals that consist only of carbs without any fiber, protein, or fat
3.) You probably also want to minimize consumption of high-glycemic index carbs, and instead consume mostly low-glycemic index foods together with (or after) fiber-rich and protein-rich foods.
4.) How many g of such carbs you can safely consume depends on how much insulin you can produce. It would probably be ideal for you to either go low-carb, or find a level of carb intake that you can maintain from one meal to another than keeps your blood glucose in the 70-140 mg/dL range 95%+ of the time.
5.) If I was you, I would monitor HOMA-%S and HOMA-%B over time, and discuss any major shifts with a doctor. If your beta-cell function drops too much, you may benefit from some exogenous insulin
Hope this helps.
Cheers
Mario
P.S.: This is not medical advise, so please make sure to discuss any issues or questions with your doctor.
Thanks for the detailed answer. I’ve been reducing my carb intake but not sure if i cash call it “low carb”. Is there are a threshold for that)
I support the idea of looking into the effects of alcohol. i noticed i can pig out on carbs without even showing a tiny spike if a have a glass of wine with it. ( whole 10” pizza + 1 glass red wine = zero spike!). Don’t know if this just delays the response but judging from just the spikes, alcohol seems like a wonder drug;) ( potential observational bias admitted)
Interesting. The literature on this clearly does not support the idea that alcohol is a wonder drug in everyone. I’ll look into this some more for sure.
Cheers
Mario
Hi Mario
Thanks a lot. What is your opinion on intermittent fasting related to diminished first insulin response?
Regards
Erik
Hi Erik,
I have several videos about this where I touch on this issue. It does seem that any fasting period longer than about 18 hours can lead to a diminished first-phase insulin response. That does not mean we shouldn’t ever fast longer; it just means we should be aware of this and re-introduce higher-carb meals slowly after fasts longer than 18 hours.
Cheers
Mario
Hi Mario – Great post! If beta-cell function is reduced (like Zoe), are there other tests you’d suggest (i.e. antibody tests) to find out why function is reduced? Could this be indicative of late onset Type 1 if the person has normal insulin sensitivity, FBG, and A1C, but has high spikes on a CGM with relatively low carb meals?
Looking forward to the post about what can be done to improve beta-cell function. Thanks!
Hi Liz,
There is a form of autoimmune diabetes than occurs later in life, and that tends to progress much more slowly than type 1 diabetes. It’s called LADA, latent autoimmune diabetes in adults. Clinical research data also suggest that LADA is under diagnosed, and that many patients with a clinical diagnosis of type 2 diabetes may actually suffer from LADA. The differentiating factor here would be the presence of certain islet autoantibodies that are also typically seen in type 1 diabetes.
All of that said, there are also numerous other reasons why beta-cell function could be reduced. Some of it is simply genetic, as annoying as that is. My suggestion is to monitor HOMA-%S and HOMA-%B as well as measures of glycemic control (fasting glucose, HbA1c, and glucose levels during the day using a CGM), and speak to a doctor about autoantibody testing only if you see a continuous worsening in these parameters that cannot be otherwise explained (for example, by weight gain, a more sedentary lifestyle, an inflammatory condition, menopause etc.).
Cheers
Mario
Hi Mario,
another great blog post to better understand the topic of insulin resistance and type 2 diabetes.
You write in the blog post:
“In this case, HOMA-%B needs to be interpreted relative to our specific level of insulin sensitivity.”
The article “Use and Abuse of HOMA Modeling” (https://pubmed.ncbi.nlm.nih.gov/15161807) also points out that HOMA-%B should not be considered in isolation.
If insulin sensitivity (HOMA-%S) is >230% and HOMA-%B is only 45%, is this an indicator of beta cell dysfunction?
Hba1c is not optimal, but normal (5.3).
CGM shows an elevation of 35-50 mg/dL postprandial in most cases.
Thank you very much, and best regards from Germany, Alexander.
Hi Alexander,
The way we would interpret this is that you are extremely insulin sensitive, but that in spite of that strong insulin sensitivity (which requires very little insulin to be produced), your beta-cells cannot produce enough insulin.
Now, that said, in that range of the graph, what you may be able to appreciate is that even small changes in fasting insulin and glucose could lead to substantial shifts in our estimates for HOMA-%S and HOMA-%B. My best guess is that if you re-did the test a few times, most of the time your HOMA-%B would be substantially higher. Otherwise, your HbA1c would be in the pre diabetic range. This exact scenario is probably the biggest weakness of this graph, simply because in everyone both fasting insulin and fasting glucose are subject to some variation. I would therefore recommend to also interpret HOMA-%S and HOMA-%B together with other measures of glucose tolerance and glycemic control, such as HbA1c and CGM data.
I will change the poster and add a second page with some information about this to help people interpret this correctly, to prevent people like you from getting overly worried where in reality they are probably fine. That said, HbA1c can also be false low, for example if your red blood cells (which carry the hemoglobin) have a short half-life. If you have CGM data, you may want to check this video here to see whether – on a moderate to high carb diet – your blood glucose levels are within the normal parameters:
The Complete Guide to Understanding Your CGM Data:
https://youtu.be/DVND90vQ0xI?si=I9mDFUmOq3n21qrn
If your glucose levels are indeed in the 70-140 mg/dL range 98% of the time, you don’t ever experience spikes to >180 mg/dL, you don’t experience hypoglycemic episodes (<70 mg/dL), and your glucose levels are back to your baseline about 2 hours after starting a meal, your glucose tolerance is perfectly fine and you can assume that your HOMA-%B measure of 45% is falsely low.
At the same time, you can probably assume that beta-cell dysfunction could be an issue for you, so you will want to prevent yourself from becoming insulin resistant in the future, and probably keep an eye on your blood glucose levels regularly.
Does that make sense?
Best,
Mario
Hi Mario, thank you very much for the detailed explanation and tips – it’s completely understandable and actionable.
I had the HOMA values calculated directly using the HOMA2 calculator.
What do you think about determining intact proinsulin as an additional marker for insulin secretion disorders?
I look forward to your next blog post. Thank you, Alexander.
Hi Mario, Thank you for your evidenced based explanations of metabolic health. For those in the range of B=25% / %S=200 can you comment on the efficacy of FMD and/or Harmine and GLP-1 agonists as interventions to stimulate beta cell regeneration (over an above life style choices embracing glycemic load reduction and positive exercise/sleep/stress protocols)?
I haven’t systematically researched where the evidence stands on these questions. It is my understanding that there is some evidence from animal experiments that a fasting-mimicking diet can lead to some improvements in beta-cell health leading to greater maximal insulin secretory capacity. However, I haven’t seen anything supporting that this is also true in humans.
The situation is similar for GLP-1 receptor agonist drugs. There is a plausible mechanism, we have some evidence in animals, and also some in humans, but it isn’t perfect evidence and certainly not conclusive. A major benefit clearly would be weight loss and specifically the loss of pancreatic fat, which is a form of ectopic fat. Whether there is an additional benefit to beta-cell function remains to be determined, I think.
Is it necessary to eat when your glucose is above 90 mg/dl? I randomly chose 90, but it could be 80 or even 70? Also, I am wondering if the that number would change as you age. When you search it on grok it returns with a “No, it is not necessary to eat when your glucose levels are above 70”, you can see all the references if you ask grok. So if it is not necessary to eat when you are above a certain level of glucose, then is it possible to fashion a diet that in conjunction with a CGM, that essentially tells you when you should eat or not eat based on glucose in the blood? Food for thought as they sa
Hi Eric,
No, it is not necessary to eat when your glucose levels in blood fall below a specific level, as long as you are normal. The body has numerous strong mechanisms to keep blood glucose in the homeostatic range and to prevent hypoglycemia (<70 mg/dL), even if you don't eat anything for several weeks.
That said, there is at least one colleague, Marty Kendall, who has proposed that we should NOT eat when our glucose levels are above a certain level because elevated glucose indicates that we already have too much fuel in our body. That is a plausible hypothesis, but his approach has not been tested in any trial I know of, so at this time it is uncertain whether this approach provides any benefits.
Cheers
Mario
Hallo Mario,
Thank you for your great video’s really helpful. My girlfriend had gestational diabetes (good subject for a future video, how) for the second time. She is lean (BMI 21) and has normal FBG, and A1C, but has relatively high spikes (140-160) early on (30min mark) on a CGM with relatively low carb meals (40grams). It looks like the first-phase insulin response is damaged.
She is now 3 months post partum and anxious about why her glucose levels haven’t restored to pre pregnancy levels (like the first time) and if that’s still possible.
We are curious about what you think can help restore beta cell functionality? Apart from all the other advise you have already given in your previous video’s (which we follow). In the previous comments you refer to possible repair strategies explained in a future video. We really could use some advise now since we worry it’s a time sensitive possibility. Could you give a sneaky preview?
Hi Sander,
Sorry I didn’t see this comment when you originally posted it. I assume it’s no longer necessary for me to respond? Hope you are enjoying the prevention program?
Cheers
Mario