It all started 22 years ago when I married Monica, who happens to have Type 1 Diabetes.
At the time, neither of us knew much about controlling blood glucose or lowering insulin levels. Unfortunately, the diabetes educators and endocrinologists hadn’t been able to help her much since she was diagnosed with T1D at 10.
Even today, we still know little about insulin. As you will learn in this article, despite the massive attention it gets, there are many open questions about this mysterious and potent hormone.
But, with two Type 1 Diabetics in the family — our son was also diagnosed with T1D when he was 17 — it’s been a topic that I’ve been fascinated by for a long time. This obsession has driven me to explore ways to lower insulin levels effectively for my family and Optimisers in our Optimising Nutrition community.
With two closed-loop CGMs live on my computer screen most of my day, I’ve had a front-row seat to how insulin and glucose respond to different foods and real-life situations. This has given me some insights and even more questions about how to lower insulin levels.
After spending a lot of time on X trying to explain my perspective on insulin, I figured it would be worth laying out my perspective on insulin in a more comprehensive article. So, in this article, we’ll look at the insulin response to:
- carbs
- protein
- dietary fat and
- body fat.
Most importantly, I’ll highlight some gaps in our current knowledge about insulin and the practical steps we can take to lower insulin levels until the many open questions are answered.
Highlights
- Carbohydrates have the largest short-term impact on insulin. However, the protein and fat in our food also elicit an insulin response, just to differing degrees over a longer period.
- Basal insulin, which is required to stop our body from disintegrating whether we eat or not, is the largest component of the insulin our pancreas produces. The larger we are, the more basal insulin we require.
- For people injecting insulin, reducing carbohydrates reduces the magnitude of the errors in estimating their insulin dose. Smaller errors are easier to correct with smaller insulin doses and tiny amounts of fast-acting carbohydrates.
- For everyone else who has a functioning pancreas, addressing the root cause of high insulin levels throughout the day—energy toxicity—with higher satiety and more nutritious food and meal choices is a more effective way to lose weight, reduce insulin total demand, stabilise blood glucose, lower insulin levels and optimise metabolic health.
Carbohydrates and Insulin Levels
Soon after we got married, we started dreaming about having kids — being a mum was Moni’s #1 dream since she was three. But we knew we first had to get serious about dialling in Moni’s glucose control to prevent all the scary complications related to diabetic pregnancies.
Fortuitously, we found an excellent doctor, Chrys Michaelis, who has a special interest in diabetes management. He paid a lot of extra care and attention, including giving us his personal mobile number to call whenever we needed help.
Over months of regular appointments, he taught us about carb counting, insulin:carb ratios, and adjusting Moni’s insulin doses to account for her changing insulin sensitivity through two pregnancies. This set us up for having two healthy babies, Jaz and Mike, who are now 18 and 19.
As a data-driven engineer, this was when I began to see nutrition as a numbers game. I figured if I could quantify nutrition and metabolism, I might be able to improve Moni’s glucose control and her quality of life.
Carb counting and insulin:carb ratios seem to be fairly common knowledge these days but for the first fifteen years of having T1D, no one in the medical system had told Moni or her parents that there was any relationship between food and insulin or how dietary adjustments could lower insulin levels. She just took two massive doses of long-acting insulin per day and then had to eat a ton of carbs throughout the day to stop her glucose crashing.
Astoundingly, the basal:bolus insulin dosing regime and the insulin:carb ratio theory, which was revolutionary for us back then and is now baked into the algorithm of modern insulin pumps, were devised by another engineer, Richard Bernstein, sixty years ago through a process of self-experimentation. Despite his best efforts, including becoming a doctor, most of his insights are still largely ignored today.
Bernstein submitted his recipe for achieving normal blood glucose levels to JAMA, but it was rejected. So, at 45, he quit his engineering job and enrolled in medical school, hoping it would give his insights more credibility and help more people.
I often imagine what Moni’s life could have been like if her parents had access to this information when she was diagnosed at 10.
Protein and Insulin
Fast-forward to January 2009, Moni got her first insulin pump. For the past 15 years, I’ve been tracking her daily insulin usage in a spreadsheet, trying to understand the factors that impact how much insulin she needs.
With the pump, not only could she skip multiple injections a day, but she could turn down her basal insulin when she was going low.
This is the major drawback of injected insulin that gets a lot of attention — we eat more when our glucose levels drop rapidly because we inject too much insulin. Because of this, insulin has developed a bad reputation as ‘the hormone that makes you fat’. But as you’ll see, it’s not that simple.
Robb Wolf, Paleo and Bernstein
In 2010, I discovered Robb Wolf, who introduced me to paleo, low-carb, and Dr. Richard Bernstein. After going low-carb, we both dropped a ton of weight.
Moni’s daily insulin requirements halved.
But when carbs are low, more of your energy comes from protein and fat, which also require insulin over a longer period. This is where insulin dosing gets more complex. Most insulins, like NovoRapid, are designed to respond to carbs over a few hours, not the longer timeframe it takes to metabolise protein and fat.
The chart below shows that the carbs cause insulin to rise rapidly, but they quickly return to baseline. In contrast, meals consisting mainly of protein and fat cause a smaller initial bump in insulin, but insulin and glucose remain elevated for much longer.
Anecdotally, we also see a similar response to blood glucose levels to high-fat meals in our Data-Driven Fasting Challenges, where people use their glucose before they eat to guide what and when to eat. While high-fat meals don’t raise glucose much, Optimisers find they have to wait a lot longer before their glucose levels return to normal.
Unfortunately, the longer-term insulin response to fat hasn’t been measured mainly because it’s smaller; it’s hard to distinguish from basal insulin in people with a functioning pancreas. However, understanding this response is crucial to lower insulin levels and manage metabolic health effectively.
Glucogenic vs Ketogenic Amino Acids
Along my journey, I learned about glucogenic amino acids and ketogenic amino acids.
Only leucine and lysine are exclusively ketogenic, meaning those above requirements for muscle protein synthesis and other critical bodily functions can be converted to acetyl-CoA and then ketone bodies.
Meanwhile, most amino acids are exclusively glucogenic, meaning, when in excess, they are converted to glucose via gluconeogenesis. Only five amino acids can be converted to ketones or glucose, depending on the body’s needs.
With this understanding, it made sense that Moni would need insulin to cover protein, only over a longer period than glucose. But maybe ‘too much protein’ was a problem and should be avoided, particularly if we’re trying to minimise insulin.
I even ran some numbers on the amino acids in different foods. I found beef has more glucogenic amino acids, while seafood contains more ketogenic amino acids. This is why many T1Ds find they need less insulin in the short term for fish than beef.
Fun fact #1: I also found that the most ketogenic protein sources were mammals from colder climates, like seal and whale blubber consumed by the Inuit. But I wasn’t sure recommending eating endangered species would be a great idea.
Fun Fact #2: Interestingly, alcohol lowers glucose and insulin in the short term. Because they are running their metabolism on ethanol, they need less glucose. Thus, alcoholics often have very low HbA1cs. Ethanol is like rocket fuel for your metabolism, with a higher oxidative priority, and thus has to be used first, before carbs, fat and protein. However, even though insulin would be lower, it’s not prudent to live on alcohol (or endangered Arctic mammals).
Type 1 Grit
It all started to come together in early 2015 when we joined the Type One Grit community.
From there, we learned how to dose enough insulin for protein with an extended bolus over five hours, with the basal insulin mainly covering the fat.
I also learned that protein requires insulin, not because it converts directly to glucose. Our steak doesn’t instantly convert to chocolate cake in our bloodstream.
Instead, protein stimulates glucagon, pushing out glucose from the liver (glycogen). Metabolically healthy people also see a commensurate rise in insulin, which keeps glucose stable after a high-protein meal. Excess protein not needed for muscle protein synthesis can be converted to glucose and stored in the liver for later use.
Protein can act as a slow-release source of glucose, which improves glucose stability compared to carbs. But if you’re relying on protein for all your glucose, you’ll probably need more dietary protein.
Many GRITers who don’t use insulin pumps use ‘regular insulin’ (or R), which acts over a longer period and is better suited for protein and fat.
The supportive community and constant display of flat-line CGMs from T1D kids helped Moni believe what was possible. She went all in with the low-carb approach. It’s hard to unsee once you experience the benefits of getting the blood glucose-insulin rollercoaster.
The Food Insulin Index – The Solution to Turning off Your Pancreas?
In early 2015, I stumbled around the Food Insulin Index, which had recently been extended to include 146 common foods (Bell, 2014).
I scraped the data from Kirstine Bell’s PhD thesis, ran a regression analysis, and found:
- protein requires about half as much insulin over the first two hours, calorie-for-calorie as carbohydrates.
- dietary fat only requires about 10% as much insulin as carbohydrates over the first two hours, and
- fructose, because some of it is converted to fat in the liver, elicits a smaller insulin response than starchy carbohydrates.
The chart below shows carbs vs the food insulin index for the foods tested. Notice how many low-carb, high-protein foods towards the left-hand side of the chart also require insulin. For more details, you can dive into the interactive Tableau version of this chart here.
Eager to find out which foods would enable Moni to need less insulin, I ranked foods based on their short-term insulin response, with fast-acting carbs being the worst and butter, cream and oil being the “best”.
I created a blog to share my new insights, which was shared far and wide by the early adopters of the quickly growing keto movement. My blog was initially called Optimising Nutrition, Managing Insulin. I shared my initial analysis with Dr Jason Fung, who shared it with Ivor Cummins, who shared it with the world.
This was when Dave Asprey’s Bullet Proof Coffee was taking off. I also listened to over a thousand podcasts with Jimmy More and his guests on the Livin’ La Vida Low Carb podcast, who regularly talked about minimising insulin and the perils of gluconeogenesis from ‘too much protein.’
After reading Jimmy Moore and Eric Westman’s Keto Clarity, I thought that if I avoided carbs and protein, fuelled with fat, and got my ketones into the ‘optimal range’, I might be able to turn off my pancreas, like my wife. With lower insulin levels, I’d be able to lose weight like a T1D who doesn’t take their insulin (also known as diabulemia).
But my body didn’t seem to subscribe to that belief. As I chugged more fat, diligently measuring my ketones, I just got fatter, as you can see from my work profile photo.
Pros and Cons of the Food Insulin Index
Eventually, I realised that living on butter and oil while keeping glucose super stable may not be the holy grail of metabolic health that I had believed.
With some more number crunching, I combined the food insulin index with nutrient density to enable Moni to stabilise her glucose, lower insulin levels, and get the protein, vitamins, and minerals she needed.
This chart illustrates the downsides of using the short-term insulin response as the primary guide to your food choices. It shows the relationship between the food insulin index and nutrient density.
As with many things in biology, this relationship follows an inverted U-shape curve, with an optimal Goldilocks zone between the extremes.
- If you inject insulin, avoiding the foods with the highest insulin index score toward the far right is smart.
- But swinging to the other extreme leads us to a very high-fat diet and avoiding the foods that contain the protein, minerals and vitamins we need to thrive and satisfy us.
In 2017, I realised most people didn’t need to minimise their insulin response to meals but rather find a way of eating to increase satiety per calorie. As shown in the next chart, foods with the lowest insulin index score also have the lowest satiety score.
While your glucose and insulin may be stable, your average glucose and insulin across the day, not to mention your body fat, might still be much higher than they should or could be.
A lower-carb diet is an excellent option, especially if you need to stabilise your blood glucose levels. People who reduce their carbs also tend to get a higher protein %, a major satiety factor.
However, if you reduce carbs AND avoid protein, you’re only managing one of the symptoms of insulin resistance and type 2 diabetes — elevated glucose — not addressing the root cause — energy toxicity. In fact, low protein with high fat is possibly the lowest satiety option, leading us to eat a lot more than we need to get the nutrients we require.
Perhaps the food insulin index’s biggest shortcoming is that it only considers the insulin response to our food over the first two hours after eating.
The food insulin index helps people on a lower-fat diet estimate the insulin demand of their food. However, it doesn’t adequately consider the insulin demand for protein and fat, which occur over a longer timeframe.
Dietary Fat and Insulin Levels
Because it’s more subtle and happens over a longer timeframe, it’s a common belief that dietary fat does not stimulate an insulin response.
Unfortunately, simple memes like this, along with “cow farts are causing global warming” and “meat causes cancer”, are simple and seductive, so they spread like wildfire but require a lot of words to unpack and address (hence the long article, sorry).
However, anyone injecting all their insulin on a low-carb diet knows that a higher-fat meal does require more insulin. The precise details of why and how much are still an open question.
To illustrate, the chart below shows insulin response curves from the food insulin index testing.
- Notice how the insulin response to glucose (black line) rises quickly and returns to baseline within two hours.
- Meanwhile, higher-fat foods like milk remain elevated at two hours, with no sign of returning to baseline anytime soon.
Unfortunately, the longer-term insulin response to fat hasn’t been measured mainly because it’s smaller; it’s hard to distinguish from basal insulin in people with a functioning pancreas. It’s usually assumed to be negligible, so it’s ignored. This leads some people to believe they can eat unlimited fat without storing it.
Young, lean and metabolically healthy people seem to be able to upregulate their metabolism to burn off the extra energy from a high-fat diet and maintain their weight.
However, our programs are full of people, particularly post-menopausal females, who have tried to keto harder and gained a ton of weight.
When they track their food, they’re surprised by how much fat they consume. They believed it was effectively free food they could eat to satiety because they were told it didn’t trigger an insulin response.
However, notice how, in the chart above, the y-axis is the change in insulin levels above baseline. It doesn’t start from zero. Instead, it’s the change above a person’s normal basal insulin levels circulating 24/7, regardless of whether they eat.
This fascinating 2021 study found that when T1Ds didn’t take their bolus insulin, a high-fat meal lowered their glucose in the first three hours. However, they saw a large rise in glucose between three and eight hours, demonstrating that some insulin is required to maintain stable glucose levels after a high-fat meal beyond the two hours that are usually measured.
The title of this 2013 study, Dietary Fat Acutely Increases Glucose Concentrations and Insulin Requirements in Patients With Type 1 Diabetes, says it all.
Finally, we have the somewhat bizarre recent study in Cell Metabolism (Johnson et al., 2024) that looked at the insulin response to each macronutrient in human pancreatic islet cells, finding that 8% of them had a larger insulin response to fat than carbohydrates. It’s hard to know exactly what to make of this study other than the fact that we still have a lot to learn about the insulin response to individual macronutrients.
Why Does Fat Raise Insulin?
It’s unclear why dietary fat raises insulin, especially since fat can easily be stored without insulin.
When we eat carbs, our blood glucose rises quickly, so the pancreas sends a signal to the liver (i.e. insulin, the messenger hormone) to tell the liver to stop releasing more energy into the bloodstream until all that excess glucose is cleared.
In contrast, the insulin response to fat is more subtle and less of an emergency. While we have limited storage capacity for glucose in our blood, liver and muscles, we have heaps of storage capacity for fat in our adipose tissue. So when your body sees a high-fat meal, there’s no fuel overflow emergency as there is with glucose. Your body says, ‘Bring it on. We can store it all in the back. We have plenty of space!’
But intriguingly, even the glycerol backbone of the fat molecule can be converted to glucose if required, which would raise insulin levels. As shown in the figure below from Lopez et al., 2020,
- carbohydrates have a direct path to raising glucose,
- the glucogenic amino acids in protein not required for essential bodily functions can be converted to glucose via gluconeogenesis,
- the glycerol backbone in the fat molecule can also be converted to glucose if required, and
- elevated free fatty acids drive increased insulin resistance because we now have more energy to keep in storage.
In the long term, the last point may be the most important but possibly most underappreciated factor.
- Once we have more fat in storage, we need more insulin to hold that fat in storage (i.e. inhibit lipolysis).
- Insulin may not be required to store dietary fat in our bodies, but once the fat is safely locked away, our pancreas needs to produce more insulin to hold it there.
- As our fat cells reach capacity, the pressure builds up, and even more insulin is required to store the fat.
- Eventually, the capacity of our fat cells taps out, and they appear resistant to the effects of insulin. We exceed our Personal Fat Threshold, and excess energy from our diet overflows into our bloodstream.
- Because our fat cells have no room left—they are now insulin-resistant—fat ends up stored in our vital organs where it doesn’t belong.
To complicate matters further, meals higher in saturated fat appear to require more insulin because they cause more insulin resistance than unsaturated fat. However, this may not necessarily be a bad thing.
A 1996 study by Holt et al. looked at the food insulin index, satiety and glucose together, concluding that meals that raised insulin and glucose the most caused greater satiety. Saturated fat may cause us to become insulin resistant sooner, which the body hears as a full signal earlier. Meanwhile, we can keep eating unsaturated fats longer before becoming diabetic.
Is it Possible to Accurately Dose Insulin for Fat?
To tackle the long-term insulin demand of fat, the Warsaw method recommends using an extended insulin bolus over 8 hours for fat-heavy meals.
You can play with the Warsaw Method calculator here. Notice how it recommends a four-hour extended bolus for high-protein meals but 8 hours for high-fat meals.
The directionality of the Warsaw Method seems to match the experience of many low-carb T1Ds, but it’s not perfect.
A 2018 study compared three methods for calculating insulin doses:
- traditional carb counting,
- the Food Insulin Index, and
- the Warsaw Method.
Unfortunately, there was no clear winner.
None of them were great.
The Warsaw Method (black line in the chart below) seemed to hit high-protein meals early and keep glucose low after eating. However, study participants experienced more hypoglycaemia after three hours. It seems the addition of the initial bolus for the carbohydrates and the extended bolus for the fat and protein is excessive.
Meanwhile, the food insulin index and carb counting methods ran participants much higher for the high-protein meals.
No Consensus on Insulin Dosing
Depending on which formula you use, you’ll get vastly different recommendations for insulin dosing. Beth McNally, a nutritionist and mother of Lachlan, an elite swimmer who happens to have T1D, shared her calculations for a low-carb meal of bacon and eggs (35g protein, 33g fat, 2.5g net carbs and 457 calories) using different formulas protocols:
- Carb counting – 0.25 units of rapid-acting insulin with the meal.
- Insulin Index – 2.2 units of rapid-acting insulin with the meal.
- Bernstein – 3.8 units of regular insulin given with the meal.
- Warsaw Method – 4.4 units of rapid-acting insulin, delivered over 8 hours using an extended bolus through the pump.
As you can see, a wide range of protocols for calculating insulin will result in a wide range of results. We still have a long way to go to reach gold standard consensus guidelines on the insulin response to protein and fat.
Trial and Error
The best approach to nailing insulin dosing for high protein and fat meals is trial and error, adjusting the timing and insulin for consistent low glycemic meals.
The slide shows the method followed by Dave Dikeman, now 18, who inspired his father RD to start the Type 1 Grit Group that has helped many T1Ds.
Bernstein’s Law of Small Numbers
For most of you who have a working pancreas, all this minutia should make you grateful that you don’t have to solve a complex math equation every time you eat; your pancreas is doing it all perfectly in the background without you knowing.
Of course, the starting point is Bernstein’s Law of Small Numbers. Smaller inputs of fast-acting carbs that require insulin cause smaller errors, which are easy to correct with small insulin doses or measured amounts of glucose to raise glucose back to healthy levels quickly.
Closed Loop Insulin Pump Systems
Thankfully, modern closed-loop insulin pump systems can do a fairly good job at mopping up errors and accounting for inaccuracies in bolus insulin dosing for fat and protein.
Some T1Ds with the latest closed-loop insulin systems can get respectable HBA1cs without telling their system what they eat. The system senses the slight rise in glucose and aggressively does insulin for it.
Unfortunately, synthetic insulin, which has a much longer half-life than the insulin secreted from our pancreas, is no match for a diet of refined carbohydrates, even with the best closed-loop insulin pump technology. To get off the glucose-insulin rollercoaster, T1Ds have to remove refined carbohydrates from their diet.
The screenshot below shows Moni’s closed-loop insulin pump in action overnight. The fat blue line at the top is the basal insulin, which continually updates based on her current blood glucose data from her CGM.
- When glucose is high and heading up, it increases the basal insulin rate to slow the release of stored energy into her bloodstream.
- When glucose is low, it lowers insulin levels and allows the release of more stored energy into her bloodstream.
Basal vs Bolus Insulin
Most low-carb T1Ds tend to focus on dosing for the protein and any carbs and then let their basal insulin account for the insulin required to cover the fat.
The chart below shows Moni’s current basal insulin profile, which I update every fortnight based on her recent glucose levels.
Across the day, her pump delivers 24 units of basal insulin, 75% of her total daily insulin dose. Some of the variation in her insulin is due to circadian patterns (e.g. to suppress glucagon associated with dawn phenomenon), but some of it also covers the fat in her meals to complement the bolus doses given for her meals.
Tidepool’s data analysis from 803 T1Ds shows that the typical adult basal:bolus split is 50:50. However, this will be higher for people on a lower carb diet. It can go as high as 90% for people on a very high-fat keto diet.
Because it’s easier to measure, the bolus (or prandial insulin, given with meals) gets most of the attention. However, for someone on a low-carb diet, their basal insulin represents the majority of their insulin demand each day.
Someone who is metabolically healthy may have a fasting insulin of 5 mIU/L, while someone who has Type 2 Diabetes and is obese will be pumping out much more insulin across the whole day to keep their body from disintegrating, regardless of their food intake.
Unfortunately, basal insulin is rarely discussed. It’s often assumed that if we keep carbs low, we effectively turn off our insulin, and dietary fat can’t be stored.
But unless you’re a T1D in the emergency ward in diabetic ketoacidosis or a fat cell in a petri dish, there will always be enough basal insulin in your system to store the fat you eat.
While we don’t yet understand exactly how much insulin we need to metabolise fat, a healthy pancreas will always produce just enough insulin to respond to the energy from your food.
Your pancreas doesn’t make mistakes.
Body Fat and Insulin Levels
After a long intro, we now come to the primary factor influencing insulin demand: body fat.
Does Insulin Make Us Fat?
Most of the time, we think of insulin as the hormone that forces energy into storage. Some have used the analogy of insulin acting like the train conductor ‘pushing’ more people onto the full train.
It’s also noted that people tend to gain weight when they start insulin therapy to manage type 2 diabetes.
This is partially true, particularly when we inject too much insulin or have too much energy in our bloodstream.
But after watching two closed-loop insulin pump systems for years, I now see insulin as more of an anticatabolic hormone that stops our bodies from disintegrating.
It’s not: eat carbs -> produce insulin -> get fat.
Unless you have someone chasing you, jabbing you in the bum with an insulin syringe, insulin is not making you fat. If this is happening to you, please call the police NOW and have them arrested!
Instead, insulin is the hormone that acts like a brake to hold energy in storage to counteract glucagon, which is like the accelerator.
Type 1 Diabetics and Insulin
Without insulin, all our stored energy, including the glucose in our liver, the fat in our adipose tissue and the protein in our muscles, flood into the bloodstream. Insulin ensures we don’t waste away, like an uncontrolled Type 1 diabetic, like the photo shown on the left around a century ago.
With poor dietary choices and inaccurate dosing, injected insulin can make us fat. But it doesn’t have to be that way.
To bring things a little closer to home, the before and after photo below is of our son Mike on his first day of getting insulin in the hospital and a month afterwards.
Five weeks after diagnosis, with 50 units of insulin per day, he added 40 kg to his deadlift, pulling 195 kg (430 lbs) in his first powerlifting competition.
To celebrate his first diaversary, after a year of protein, training and insulin, he set an under-18 world record, pulling 245 kg (540 lbs) or three times his body weight!
I could go on with heaps of examples of people with type 1 diabetes who are thriving with way better body composition than average. They don’t fear insulin. They eat well, exercise, and take as much insulin as required to keep their glucose in the normal, healthy range.
Insulin Plummets During Weight Loss
When we eat less, insulin lowers and lets glucagon push more energy into circulation.
The chart below shows Moni’s daily insulin requirements during our first Macros Masterclass in January 2019.
Before the challenge, she was already low carb with plenty of protein and fat to maintain her weight. During the challenge, she dialled back her dietary fat while maintaining the protein. Her insulin requirements plummeted from 25 units per day down to a low of 13 units per day! We both lost over 10% of our starting body weight during the challenge.
While the rise in insulin in the two hours after you eat is mainly related to the amount of carbohydrates you eat, the total insulin produced by your pancreas is a function of the energy in your diet AND the energy stored in your body.
The more fat you store in your body, the more insulin you need to be to hold all that energy in storage. Regardless of whether you inject it, the bigger you are, the more insulin you need.
Measuring insulin by merely focusing on the carbs in your diet is a bit like trying to measure the volume of the ocean by measuring the height of the waves when you’re standing on the beach.
Insulin: Friend or Foe?
Insulin gets a lot of bad press and is blamed for many bad things.
But what if insulin is like a firefighter who happens to be at the scene?
What if insulin is not the bad guy but trying to save us from the energy crisis our nutrient-poor, low-satiety diet has caused?
Your pancreas knows that having massive amounts of energy in your bloodstream is dangerous, so it’s just working to save you by holding the energy back in storage until you need it later when the famine comes or you find a way of sustainably eating less.
If you’re battling insulin resistance or type 2 diabetes, reducing carbs enough to achieve healthy glucose levels (i.e. a rise of less than 30 mg/dL or 1.6 mmol/L after eating) is a great first step.
But the next step is to attack the root cause of the majority of your insulin — our excess stored energy.
For more details of what this looks like in practice, see:
- Avoid These 2 Diet Danger Zones for Increased Low-Carb Satiety
- How to Optimise Your Low-Carb Macros for Satiety and Weight Loss
Towards a Better Understanding of Insulin
If you have a fully functioning pancreas, hopefully, this article will give you some appreciation for how wonderful insulin is and the complex math your pancreas does 24/7 to manage the fuel level in your bloodstream precisely.
Perhaps in a few years, we’ll have continuous insulin monitors available so everyone else can see how their pancreas responds to food. If this ever happens, I think most people will quickly realise that the real problem we need to address is not insulin but energy toxicity.
Once we understand that it’s not just carbs that impact insulin, we realise that the solution is not swapping carbs and protein for fat but rather eating in a way that nourishes our body and provides greater satiety.
However, for the growing number of people who need to know precisely how much insulin to inject, I think much more work still needs to be done. The optimist nerd in me dreams that it’s a problem that could be solved with more data.
But what could this look like?
Closed-loop insulin pump algorithms track carbs on board (COB) and ensure enough insulin is available to achieve the target blood glucose in a few hours. The crucial factor here is not just the amount of carbs but the assumptions around how long it takes for carbs to be metabolised and no longer require insulin.
But what if we also tracked protein and fat on board with a similar decay profile?
What if they could also track the other things in our food that influence insulin and glucose in unique ways, like:
- different types of carbs (e.g. fibre, starch, fructose, glucose, lactose, maltose and sucrose),
- different types of fats (e.g. saturated fat, monounsaturated fat, polyunsaturated fat, omega 3, omega 6, cholesterol)
- amino acids (e.g. alanine, cysteine, histidine, isoleucine, leucine, lysine, methionine, phenylalanine, threonine, tryptophan, tyrosine and valine) and
- alcohol.
Imagine if you could enter your food in an app like Cronometer (or even just take a photo of your meal), which would send all the macronutrient and micronutrient data to your closed-loop insulin pump system and model how much of each energy source and nutrient was still left to metabolise.
With enough data (it would take a lot of crowdsourced data), we could tease out the timeframe each component of our food takes to be metabolised and then fine-tune that for each person injecting insulin based on their trained dataset. This deeper insight into our insulin response to food could also inform the development of future clinical guidelines.
We could also reverse engineer this data to help people understand which foods would nourish their bodies and optimise their insulin treatment to align with their unique context and goals.
This is an ambitious dream, but the technology to make it a reality is getting closer. It might be a few decades away. Maybe it’s a problem my son will solve after I’m gone? He’s smarter than me and eager to be part of the solution.
While this task is immense, I’m optimistic. I’ve been amazed to see what the DIY T1D community has managed to achieve because they are so motivated and committed. They are not waiting for the mainstream medical and nutrition institutions to save them. They want the solutions so they or their children can thrive now!
Action Steps: Practical Tips to Lower Insulin Levels
This article has discussed a ton of technical details about insulin.
However, there are a few practical takeaways and tools for those who want to reduce their insulin and stabilise their blood glucose levels.
After a decade of trying to help my family and the Optimising Nutrition Community make more informed food and meal choices, this food search tool is the latest and greatest.
For those Injecting Insulin
If you inject some or all your insulin, the first crucial step is to get off the blood glucose insulin rollercoaster by reducing the carbohydrates that require more insulin in the short term.
To find foods that will help you do this, use carbohydrate sliders to identify foods that contain less carbohydrates per serving.
If you mouse over the foods, you’ll also see additional details that will help you make more informed food choices:
- Glycemic index: GI ranks foods from 0 to 100% based on their impact on blood sugar levels. Choose lower GI foods to reduce the rise in glucose after you eat.
- Glucose score: The area under the curve glucose responds to foods over two hours after eating them.
- Insulin index: A ranking based on the insulin response to food in the two hours after eating it for equal calorie portions of food.
- Insulin load: The food insulin index multiplied by the number of calories in a typical portion of that food.
For the Rest of Us
But for the majority of us who want to lose weight, optimise our metabolic health and feel satisfied with our food:
- PLEASE STOP WORRYING ABOUT INSULIN.
- Unless you inject all your insulin, you can’t measure it, and it’s way too complex to micromanage.
- Insulin is not the bogeyman that MAKES US FAT. It’s a distraction.
Instead, let’s focus on the root cause of high insulin levels — ENERGY TOXICITY — for which the solution is higher satiety foods and meals.
Trust me, once you find a way of eating that nourishes your body and satisfies you, you’ll eat less, lose weight, and your hard-working pancreas will get a well-deserved holiday.
- Foods towards the left of this chart will drive us to eat more, raising our insulin levels in the short and long term.
- Meanwhile, foods towards the right will empower us to eat less, lose weight and lower our insulin levels throughout the day.
- For bonus points, target foods toward the top that provide more nutrients your body craves.
Before you go, click here to open the interactive food search tool and surf for new foods that align with your goals.
Marty I am not diabetic but now I truly appreciate the grueling hours of work and research, tweaking up and down insulin and which macros your wife and son need to thrive and live their best. I loved this article. You are a beautiful person.
Thank you so much! It’s fun to be able to share my learnings with the world.
This was the most timely article to read. I attend a Diabetes info session last night, and the dietician told the people that every thing we stuck in our mouths contains carbs. Fat, meat, eggs, fruit, veggies etc. I was listening to this and wondered where will I start looking for RD’s posts commenting that prot that does not turn into cake. I will have to print it and write out the difficult terms to be able to talk about it.
Thanks for sharing this, Marty! I’ve been through your courses and they deliver quite the education. Question:
What are your thoughts on the Fasting Mimicking Diets?
I’ve seen great results; so much so that I’ve recommended Prolon to clients and wrote a Prolon 1 Day Reset Review about it here:
https://jasonryer.com/prolon-1-day-reset-review/
Let me know your thoughts – and thanks again for all you do!