The Paradox of Low-Carb Diets: A1c vs. Metabolic Health

The conflict between glucose levels and metabolic health is unique for T1Ds

As we all know, excess blood glucose can cause microvascular complications (blindness and kidney disease); harm arteries and give rise to cardiovascular disease; or interfere with the immune system, making it hard to fight disease; or aid in clearing cholesterol; or grow new cells, or repair damaged tissues, think, breathe, everything.

Sadly, insulin can do the same thing—and in many cases, worse and more quickly—if dosing levels are too high. (For more, see my article, Insulin and Heart Disease: What You Should Know.

For this reason, many advocate for a “low-carb diet”, as it not only reduces blood glucose levels, but also requires considerably less insulin.

Sounds like a magic formula right? Well, it’s not that simple.

The most important thing to remember throughout all of this is that metabolic fitness is the ultimate metric for health, and lower A1c levels does not, by itself, confer metabolic health. One can have low or normal blood glucose levels, while also being very unhealthy metabolically.

Ironically, one can also have high glucose levels while also being metabolically healthy.

One illustration comes from Larry Distiller (Principle Physician and Managing Director at the Centre For Diabetes And Endocrinology in Johannesburg), who conducted a comprehensive literature review called, “Why do some patients with type 1 diabetes live so long?” His review paper cites multiple studies that those who lived the longest were shown to have excellent metabolic and cardiovascular health, despite having elevated glucose levels. One article included in his review is the “Golden Years Cohort,” which includes 400 T1Ds who’ve had the disease for over 50 years. Their mean HbA1c was 7.6% (± 1.4), with some levels as high as 8.5%-9%. None had an HbA1c below 7%.

Interestingly, among those who happened to die early had significant problems with lipids (high cholesterol), kidney disease, and other factors, any of which could be exacerbated by high glucose levels, but the correlation was weaker. That is, they showed signs of significant metabolic stress independently of A1c levels, with many having A1c levels <7%.

Elevated glucose levels are bad. But an unhealthy metabolic system is far worse.

And this is what brings us to diets: Aiming for diets that solely focus on lowering glucose levels without also considering metabolic health is the trap that some can get into. The danger of chasing glucose progressively lower with an inappropriate diet—one that does not comport with your genetic disposition—is that it can introduce severe metabolic stress.

This is a highly dense and complex topic, and as such, it is controversial for very good reasons: There are pros and cons to both low-carb diets and high-carb diets, particularly when it comes to T1D health and management.

Let’s begin by understanding what we’re talking about in the first place.

What is a Low-Carb Diet?

The term "low-carb diet" is not universally defined, and this ambiguity complicates both research comparisons and practical recommendations.

One of the most familiar low-carb diets is the Bernstein diet, which he describes in his 1977 book, Dr. Bernstein's Diabetes Solution: The Complete Guide to Achieving Normal Blood Sugars. In essence, Bernstein recommends no more than 30g of carbs per day. (A longer summary and analysis can be found in the article, “What Is Dr. Bernstein's Diabetes Diet?”) Dr. Bernstein’s approach—recommending “no more than 30g/day” of carbohydrate—is indeed highly restrictive and does not scale for body weight or physical activity.

Less well-known, but is gaining attention, is the “Mastering Diabetes” movement, led by Cyrus Khambatta, PhD (nutritional biochemistry from UC Berkeley) and Robby Barbaro, MPH—both living with T1D themselves. Their book is called, Mastering Diabetes: The Revolutionary Method to Reverse Insulin Resistance Permanently in Type 1, Type 1.5, Type 2, Prediabetes, and Gestational Diabetes, calls for a low-fat, plant-based, whole-food diet that includes very high carb loads—up to 700g carbs a day.

In between these two extremes is where the scientific literature and international guidelines land. The article Carbohydrate Restriction in Type 1 Diabetes: A Realistic Therapy for Improved Glycaemic Control and Athletic Performance? offers the following operational definitions:

The idea of associating one’s weight with a number of carbohydrates was shown to be useful in a study described in the paper, Effects of a low-carbohydrate diet in adults with type 1 diabetes management: A single arm non-randomised clinical trial. Here, scaling carbohydrate targets by grams/kg body weight depending on activity level seemed optimal for gauging one’s balance between energy consumption and expenditure. Values range from 3–5g/kg for low-activity up to 12g/kg for elite/extreme activity. More intensive physical activity requires direct access to glucose; hence the carb ratio is higher.

As you can see by this table, there’s a lot of variability in what constitutes each tier of carb intake. The fact that most medical literature recognizes 130g/day as being a “low-carb diet” may surprise many.

Unless you’re aware of these definitions, you don’t actually know if you’re on a low-carb diet. Most people don’t count carbs, and among who do, few are very good at it. For this reason, many think they’re on a low-carb diet, when they’re not. And vice versa.

For some, 130g/day feels like a lot, and certainly Bernstein followers would say this is not low-carb. But it is. And the ADA thinks so too.

Then there’s the psychological aspect: Many think they’re on a low-carb diet but aren’t, only because they’re not eating as many carbs as they WANT. There’s no question that restricting your food intake is hard, nor is it restricted to T1Ds, as is well known. But just because you may be struggling—and succeeding—at avoiding those high-carb foods you love, it doesn’t mean you’re on a low-carb diet.

Regardless of whether any given diet constitutes “low carb” in your own mind, the idea of a low-carb diet is to emphasize fat-burning for energy, and to reduce carb intake in order to keep insulin levels low and reduce high glucose excursions.

While a great idea in theory, this complex metabolic process is highly variable among individuals. And that’s why guidelines are have ranges. It’s important to take into account body weight and activity level. A small, 110 pound sedentary person would need far less carbs than a 220 pound, lean marathon runner.

These definitions are useful if we’re going to evaluate their health benefits, or potential harms.

The High-Carb Counter-Movement

As noted earlier, the “Mastering Diabetes” movement argues that while low-carb diets may improve short-term blood glucose control, such diets increase long-term risk for chronic diseases including cancer, high blood pressure, high cholesterol, chronic kidney disease, and fatty liver disease. These are the byproducts of poor metabolic health.

The numbers they report are striking: Barbaro's insulin-to-carb ratio is reportedly 1:58, eating 600-700g of carbs daily while taking only 26 units of total insulin—a level of insulin sensitivity that seems almost impossible by conventional T1D standards.

The Mechanistic Debate

Their argument centers on insulin sensitivity rather than carb restriction. They claim that dietary fat, particularly saturated fat, causes intramyocellular lipid accumulation that blocks insulin signaling—essentially that fat makes you insulin resistant, requiring more insulin for any given amount of carbohydrate. By drastically reducing fat intake (to 20-30g daily), they argue you can achieve such high insulin sensitivity that massive carbohydrate loads require minimal insulin.

This creates an interesting paradox: Both ultra-low-carb and ultra-high-carb approaches can achieve excellent A1c levels, but through completely opposite mechanisms—one by avoiding glucose spikes altogether, the other by maximizing the body's ability to process glucose efficiently.

What both extremes inadvertently demonstrate is that the human metabolism—even in T1D—has remarkable flexibility. Some individuals achieve low glucose levels by becoming efficient fat-burners (ketogenic adaptation), while others achieve it by becoming highly insulin-sensitive carbohydrate processors.

But again—never forget—it’s not just about low glucose levels. We need to measure metabolic health. Indeed, the benefits of each may be limited by genetics. Some just may have genes that lean towards one, or the other, or both, or neither.

The tragedy is that both camps often claim theirs is the “only” way, when the evidence increasingly suggests that optimal diet depends on individual genetics, lifestyle, activity level, gut microbiome, and numerous other factors we're only beginning to understand.

The high-carb diet has not been around long enough to investigate its long-term health outcomes, but low-carb diets have, so let’s take a look at what studies have shown.

Studies supporting low-carb diets

Advocates for low-carb diets frequently cite three particular studies that warrant close examination.

The first is a paper published in 2018 by Dr. Bernstein et al titled, Management of Type 1 Diabetes With a Very Low-Carbohydrate Diet, published in the journal Pediatrics.

What this study is and what it found

This 2018 Pediatrics paper surveyed members of an international social media community of people with T1D (and parents of children with T1D) who already follow a very low‑carbohydrate diet (VLCD). Key reported features from the abstract:

• Mean carbohydrate intake: 36 ± 15 g/day.

• Mean HbA1c: 5.67% ± 0.66% (exceptionally low for T1D).

• Reported diabetes-related hospitalizations in the prior year: 2%, including 1% for ketoacidosis and 1% for hypoglycemia.

• Duration of VLCD: 2.2 ± 3.9 years.

• “Suggestive evidence” of T1D was obtained for 86% (273/316), based on a 3‑tier internal scoring rubric.

The authors appropriately conclude that “the generalizability of these findings requires further studies, including high-quality randomized controlled trials.” Indeed, those trials (and other population-wide data) have been conducted and published more broadly, discussed in the next section.

Core limitations that constrain interpretation

• Severe selection bias (enrichment for success cases): Participants were recruited from an online community specifically dedicated to VLCD for T1D. By design, this draws people who have chosen, persisted with, and often succeeded on the approach. Individuals who tried VLCD and stopped due to difficulty, adverse events, or poor outcomes are underrepresented.

• Self-reported outcomes and exposures: HbA1c, hospitalizations, insulin dosing, and adverse events were primarily self-reported. While the abstract notes “confirmatory data” from providers and records, it does not quantify how many participants had objective verification nor the concordance rates. Without rigorous, uniform validation, recall and reporting biases remain likely.

• Cross-sectional survey without a control group: The study captures a snapshot of current outcomes in adherent VLCD users. It cannot establish causality or even strong associations relative to standard carbohydrate intake, because there is no contemporaneous comparison cohort matched on age, diabetes duration, technology use, socioeconomic status, or care setting.

• Non-standard T1D case verification: The “suggestive evidence” scoring rubric for T1D diagnosis blends age/weight at diagnosis, autoantibodies, insulin requirement, and clinical presentation. That leaves room for misclassification and heterogeneity (e.g., LADA, ketosis-prone diabetes, atypical phenotypes), particularly in an online survey setting.

• Adherence/survivorship bias: Mean VLCD duration of 2.2 years implies respondents are those who sustained the diet. Outcomes in the first months (when adverse events or dropout could be higher) are not captured; those who discontinued likely did not respond.

• Pediatric data primarily via parent report: Some outcomes for children (e.g., hypoglycemia recognition, growth metrics, psychosocial impact) are difficult to assess accurately via parent surveys alone. Growth and development data are notably absent.

• Confounded by contemporary diabetes technology and care: The study year (2018) sits at an inflection point in CGM and pump adoption. High CGM use, structured insulin protocols, or frequent clinician follow-up could drive excellent HbA1c independent of diet. The abstract does not report CGM metrics (time in range, time <70 mg/dL), device use, or education intensity.

• Diet quality and metabolic safety not detailed: Beyond grams of carbohydrate, there is no granular macronutrient pattern (fat types, protein amounts), ketone monitoring, lipid profiles, micronutrients, or data on euglycemic ketoacidosis risk (especially relevant if any SGLT2 use). For children, growth velocity and pubertal progression are crucial safety endpoints not discussed.

• Conflict of interest and advocacy alignment: Several authors are publicly associated with low-carb advocacy or industry (e.g., Bernstein, Hallberg). COIs do not invalidate data, but they increase the need for independent replication and rigorous methodology to avoid confirmation bias in recruitment, analysis, and interpretation.

Collectively, these limitations explain (and justify) the authors’ caveat about generalizability and the explicit call for randomized controlled trials.

Countervailing concerns

• Risk of ketosis and euglycemic ketoacidosis under stressors (infection, missed insulin), especially if SGLT2 inhibitors are involved.

• Lipid profile changes and long-term cardiovascular risk depending on fat sources.

• For children, growth and psychosocial impacts demand careful monitoring.

• Sustainability and quality of life vary widely; adherence can be challenging.

Plausibility does not substitute for controlled evidence. The signal in this survey is hypothesis-generating, not practice-changing by itself.

Again, the next section will review a much larger list of papers that do not suffer these confounding factors.

The FASTER study

Another of the more frequently cited papers is the FASTER study (Volek et al., 2016), which showed that elite ultra-endurance athletes on ketogenic diets maintained normal muscle glycogen levels through gluconeogenesis. Low-carb advocates trumpet this as proof that carbohydrates aren't necessary for athletic performance.

• Twenty elite ultra-marathoners were studied - 10 on high-carb (59% carbs) and 10 on low-carb (10% carbs) for an average of 20 months.

• Peak fat oxidation was 2.3-fold higher in the LC group (1.54 vs 0.67 g/min).

• Despite marked differences in fuel use, there were no significant differences in resting muscle glycogen and the level of depletion after 180 min of running

However, several critical factors undermine this conclusion for T1Ds:

1. Population Specificity: The study examined 20 elite ultra-marathoners who had been keto-adapted for an average of 20 months. None were T1Ds. These athletes represent extreme metabolic outliers—people whose bodies have adapted over years to running 50-100+ miles at a time. Extrapolating from this population to typical T1Ds is like studying Formula 1 drivers and applying the findings to teenage driving students.

2. No Performance Metrics: Crucially, the FASTER study didn't measure whether the low-carb athletes actually performed better—only that they maintained glycogen differently. In fact, subsequent research has shown that while fat-adapted athletes can maintain submaximal efforts, high-intensity performance consistently suffers. A 2017 study of elite race walkers found that ketogenic diets impaired their competitive performance despite increased fat oxidation.

3. The Gluconeogenesis Problem for T1D: The study's finding that glycogen was maintained through gluconeogenesis (making glucose from protein and fat) highlights a critical issue for T1Ds: This process requires precise hormonal orchestration involving glucagon, cortisol, and growth hormone—systems that are dysregulated in T1D: They have impaired glucagon response due to beta cell destruction, making efficient gluconeogenesis unreliable at best.

4. The Exercise Intensity Limitation: The FASTER study tested athletes at 64% VO2max—a moderate intensity where fat oxidation is maximized. But real-world exercise involves much higher levels of intensity, particularly for improving metabolic fitness. When intensity exceeds ~70% VO2max, even fat-adapted athletes must rely primarily on glucose. And you can’t build or strengthen mitochondria unless you exceed the “lactate threshold” (70% V02max). Even worse for T1Ds without reliable glycogen stores or gluconeogenesis, this creates a dangerous situation where hypoglycemia becomes likely during any intense effort.

In other words, metabolic health is not conferred or even hinted at by this study. It merely demonstrated that some individuals are able to efficiently utilize non-glucose substrates for fuel.

The TypeOneGrit Survey

The 2018 survey of 316 TypeOneGrit community members following Bernstein's diet showed impressive average A1cs of 5.67%. However, this study exemplifies nearly every form of bias possible:

• Survivor bias: Only those successful enough to remain in the community were surveyed.

• Reporting bias: Self-reported data from highly motivated participants.

• Socioeconomic bias: 84% college-educated, 88% white—a population with exceptional resources.

• Attrition invisibility: No data on the countless individuals who tried and abandoned the approach.

More tellingly, when researchers attempted a follow-up study of the same community, only 8.9% of previous participants responded—suggesting that even among this highly selected group, long-term adherence may be problematic.

This is basically it—there are no long term, epidemiological studies, or randomized control trials that have shown metabolic benefit to an ultra-low-carb diet.

The Mechanistic Mismatch

The theoretical framework for low-carb diets in T1D rests on avoiding glucose spikes by avoiding glucose itself. While logical on its surface, this approach ignores fundamental physiology:

1. Glucose is Essential: The brain requires ~130g of glucose daily (typically, 12g/hour). While ketones can substitute for some of this, complete replacement isn't possible. The body will make glucose through gluconeogenesis regardless, but this process is energy-expensive and produces metabolic stress, which produces free radicals (oxidative stress) that is directly linked to cardiovascular disease. (And that’s just the brain. Muscles are the largest sink of glucose.)

2. Protein Becomes Glucose: On very-low-carb diets, 50-60% of protein converts to glucose via gluconeogenesis. This makes protein dosing unpredictable and can cause delayed hyperglycemia 3-5 hours after meals.

3. Exercise Becomes Dangerous: Without adequate glycogen stores, exercise capacity is limited to low-intensity efforts. Any spike in intensity—running for a bus, playing with children, escaping danger—risks severe hypoglycemia. Yes, some people can do this well, but they are genetic outliers.

The best data to truly assess health are long-term studies that study populations of people, especially involving greater diversity of behaviors and outcomes. Ideally, RTCs (random control trials) that involve individuals who participate without pre-existing bias are best candidates. Over the ensuing decades, low-carb diets, including ketogenic diets, have been put to the test in thousands of clinical studies and trials, leading to significant refinements about what is considered “healthy,” especially for T1Ds.

Long-term Health Outcomes of Low-Carb Diets

While low-carb diets help reduce A1c levels, the studies also revealed the potential drawbacks. In the 2019 comprehensive literature review, “Low-Carb and Ketogenic Diets in Type 1 and Type 2 Diabetes,” over 100 peer-reviewed papers, studies, and books that spanned hundreds of thousands of people over decades paint a constellation of risks. Highlights include:

• Those on low-carb diets generally have significantly higher levels of low-density lipoproteins (LDL) substantially increasing the risk of cardiovascular disease, according to a randomized control trial published in the journal Nutrients. The elevated LDL levels are primarily due to a unique metabolic adaptation to the lower carb intake, not just “eating the wrong foods”. When carb intake is severely restricted, insulin levels remain chronically low, and since insulin is the primary hormone that signals the body to store fuel, two things happen: Adipose tissue releases free fatty acids (FFAs) into the bloodstream as the primary energy source, and the liver must process and package this high volume of circulating and dietary fat for delivery to the muscles and organs that need fuel. The liver packages fats and cholesterol into Very Low-Density Lipoproteins (VLDL) for transport. These VLDLs are broken down in the bloodstream, eventually becoming Low-Density Lipoproteins (LDL), which is the primary driver of cardiovascular disease.

• Children on low-carb diets are more prone to anthropometric deficits, and an increase in the risk of eating disorders, especially among adolescents, according to the 2025 literature review, Low-Carb and Ketogenic Diets in Type 1 Diabetes: Efficacy and Safety Concerns.

• T1Ds on low-carb diets statistically suffer higher rates of hypoglycemia, as well as euglycemic DKA (diabetic ketoacidosis with normal glucose levels), due to insufficient glycogen stores.

• Compliance rate is low, with 70-80% failing to adhere to the diets after two years, triggering a severe rebound effect, where their T1D management was worse than before attempting the diet. 

• The psychological burden of managing both T1D and a highly restrictive diet has been found to lead to mental health disorders.

More importantly, the cited literature review paper reveals that nearly all T1Ds on low-carb diets are metabolically unfit, even among those whose average glucose levels were at or below recommended targets, leading many to die prematurely from all-cause mortality. As the analysis showed, T1Ds on low-carb diets had a 32-50% increased risk compared to those following moderate carbohydrate intakes.

This is yet another reminder to assure your periodic blood work includes the metabolic panel and fractionated lipid panel, regardless of what diet you’re on.

In the literature review, Low-carbohydrate diets and all-cause mortality: a systematic review and meta-analysis of observational studies, people with the lowest carbohydrate intake (quartile 4 of LCD)—which includes the Bernstein diet—had the highest risk of overall mortality (32% increase), cardiovascular disease mortality (50% increase), and cerebrovascular mortality (51% increase).

In another literature view, The low-carbohydrate diet and cardiovascular risk factors: Evidence from epidemiologic studies, researchers found an association between low-carbohydrate diets and mortality was stronger in non-obese participants (48% increased risk) than in obese participants (19% increased risk). Importantly, a low-carbohydrate dietary pattern accompanied by high intake of animal-based protein and fat was associated with higher all-cause mortality.

An article in the ADA Journal, Diabetes Spectrum, titled, Low-Carbohydrate and Very-Low-Carbohydrate Diets in Patients With Diabetes cites the potential short-term benefits do not outweigh the long term risks. This is why ADA meal plans start around 130g carbs/day for T1Ds, according to their article, Conquer Your Carb Confusion. As noted above, this level meets the guideline for “low-carb diet”, which is why some people point to the ADA as advocates of “low-carb”, even though it’s not advocating for the Bernstein level of 30g/day.

The Genetic Component

This raises an interesting question: Given all the studies showing how dangerous low-carb diets are, what accounts for those who apparently do so well with them?

Dr. Bernstein himself lived past 90, exercised well, and so on. True. But other low-carb advocates didn’t. Two of the most famous cases are Robert Atkins (the famous low-carb “Atkins diet”), who died of a heart attack, and Charles Poliquin (a prominent strength coach and low-carb advocate who promoted the "meat and nuts" diet) died at 57, also from a heart attack.

Indeed, we can look to the stats on all-cause mortality to find even more unhappy customers.

But that doesn’t mean people who are currently benefiting aren’t interesting to study. Indeed, what accounts for those who do well with low-carb diets?

The leading explanation is simply genetics. Recent research shows that some individuals are metabolically “suited” to extreme carbohydrate restriction OR excess, simply because their genetics allow it.

Nutrigenetics studies how a person's genes affect their response to nutrients in ways that lead to different metabolic responses to carbohydrates. One can have genes that are detrimental to health, and other genes that are protective. One can have genes of both—where they are at higher risk for something, but also have a gene that’s protective.

The constellation of genes that drive these things combine in ways that can sometimes work very well together, and other times conflict with one another. This is why each of us is somewhat unique. But when you look at populations as a whole, we see people distributed into pockets, something like the 80/20 rule. That is, 80% of people are similar to one another in most ways, while the remaining 20% are outliers. Mix all these genes up and you get some really crazy outliers.

For example, there’s a phenomenon called, Metabolically Healthy Obesity (MHO), referring to those whose weight far exceeds what is considered “obese”, yet experience no evidence of any metabolic syndrome. According to the JAMA article, Trends in the Prevalence of Metabolically Healthy Obesity Among US Adults, 1999-2018, the proportion of obese people who were metabolically healthy is roughly 15.0%. (Interestingly, the rates are far higher in Europe, with English women topping the list at 30%.)

To figure out what’s right for you, one must begin with two seemingly contradictory ideas: First, that you don’t really know what you’re like unless and until you try something; second, you’re far more likely to be like 80% of the population, so that’s probably the best place to start your journey.

Given the propensity of harm experienced by the majority of people who attempt ultra-low-carb diets (e.g., Bernstein), the statistical likelihood among those who thrive suggests that they have unique combinations of genes that are not necessarily rare on their own, but in their unusual collective aggregate, they are statistical outliers. It may well be that such people can and should be on very low carb diets. But just because they can, doesn’t mean others should. Indeed, it’s highly unlikely that others will succeed.

Those who promote most ultra-low-carb diets would be similar to those few Metabolically Healthy Obese people saying that they eat anything and everything, and they’re perfectly healthy, so you should try it too.

We can all see how preposterous that would be. Indeed, consider the opposite: Studies have shown that weight loss in people who are metabolically healthy obese (MHO) can lead to negative health outcomes. The article, Effects of Weight Loss Among Metabolically Healthy Obese Men and Women looked into this and found:

Weight loss among metabolically healthy obese (MHO) individuals characterized by low cardio-metabolic risk and low prospective risk of type 2 diabetes and cardiovascular disease may be unnecessary and paradoxically may actually increase health risk.

For those whose obesity is a natural, healthy state for their body type, losing weight causes their metabolism to react as though they are in a state of famine. They may look like a healthy, lean, normal person, but their metabolic health is in a state of serious deterioration, such as drops in insulin sensitivity, an increase in blood pressure, and other signs of metabolic distress. Essentially, their bodies fight against the weight loss, and in doing so, they lose the very metabolic health that made them unique.

In the same way, just because you have low A1c levels and can appear to convert non-glucose substrates into energy may appear superficially that you’re healthy, but you could very well be suffering metabolically. And this is precisely the finding that decades of medical literature (cited above) has found.

In fact, the same is true of the high-carb advocates as well. It works for a small percentage of people, but it’s not something that should be promoted to the wider population.

The reality is that picking the right diet is entirely related to your metabolic need, which is primarily governed by your genetics, which not only sets boundaries about the limits you can tolerate, but also tilts towards what your mitochondria can actually do.

When examined critically, the evidence for low-carb diets in T1D supports a much more limited conclusion than advocates claim:

• May benefit: Sedentary individuals with significant insulin resistance who cannot or will not exercise.

• Possibly appropriate: As a short-term intervention (3-6 months) for weight loss under medical supervision.

• Clearly contraindicated: For children (growth impairment), athletes (performance degradation), and those with history of eating disorders.

This now brings us full circle to why all of this is important: Metabolic Health. It’s not all about A12c levels. Carb intake is necessary to do activities that improve metabolic health. That should be your ultimate priority.