Right to Measure Insulin

“Those with cardiovascular disease not identified with diabetes are simply undiagnosed.” — Joseph R. Kraft, Diabetes Epidemic and You, 2008

The hormone medicine looks past

Insulin is the master anabolic hormone of the human body. It gates the entry of glucose into cells, directs the storage of fat, restrains the breakdown of adipose tissue, activates the mTOR pathway that governs cellular growth, modulates sodium retention in the kidney, and shapes the endothelial biology of every artery. When its signalling is deranged, the derangement does not stay in one organ. It shows up as hypertension, dyslipidemia, central adiposity, non-alcoholic fatty liver disease, polycystic ovary syndrome, reactive hypoglycemia, cognitive dysfunction, and eventually type 2 diabetes. The list is not a coincidence. It is the clinical signature of a single upstream event: chronic hyperinsulinemia.

For a hormone with this reach, medical routine measures insulin astonishingly little. A person can accumulate a full complement of insulin-resistance markers over twenty years, be treated in parallel for hypertension, high triglycerides, and abdominal weight, and never have their insulin measured once. The variable at the centre of the pathology is, in the ordinary clinical encounter, invisible.

This is not a knowledge gap in the research literature. Kraft published in 1975. Reaven named the syndrome in 1988. DeFronzo formalised the clamp technique in 1979 and gave the ADA’s Banting Lecture on insulin resistance in 1988. The mechanistic work of Shulman and Petersen at Yale on intramyocellular lipid and mitochondrial dysfunction dates from the late 1990s. The pathophysiology has been mapped in detail for four decades. What has not caught up is the clinic.

A history that never became routine

Joseph R. Kraft (1933–2017), a pathologist at St. Joseph Hospital in Chicago, spent thirty years performing what would eventually total over 15 000 oral glucose tolerance tests with simultaneous insulin measurement at 0, 1, 2, and 3 hours. Where the standard OGTT reads only glucose and delivers a binary verdict of “normal” or “diabetic,” Kraft’s five patterns read the insulin dynamic. Pattern I is the physiological response: modest rise, timely fall. Patterns II, III, and IV are progressive hyperinsulinemic responses in which glucose still looks acceptable while insulin secretion is escalating catastrophically. Pattern V is frank insulin deficiency.

Kraft’s finding was blunt: in the population he tested, roughly 75% of people with “normal” glucose curves already had a pathological insulin curve. In his own formulation, cardiovascular disease without a diabetes diagnosis is not the absence of diabetes. It is undiagnosed diabetes, caught on the glucose axis alone, missed on the insulin axis where the dysfunction had been visible for years.

Gerald M. Reaven (1928–2018), of Stanford, delivered the Banting Lecture in 1988 and introduced the concept of Syndrome X: the clustering of insulin resistance, compensatory hyperinsulinemia, glucose intolerance, elevated VLDL triglycerides, reduced HDL cholesterol, and hypertension. He argued, correctly and decades ahead of guidelines, that these were not five diseases but one, and that insulin resistance was the shared upstream driver.

Ralph A. DeFronzo, of the University of Texas Health Science Center at San Antonio, developed the hyperinsulinemic-euglycemic clamp in 1979, still the gold standard for quantifying insulin sensitivity in research. His 1988 Banting Lecture, published as “Lilly Lecture 1987: The Triumvirate: β-Cell, Muscle, Liver. A Collusion Responsible for NIDDM”, laid out the three-organ pathophysiology that remains foundational.

Gerald I. Shulman and Kitt F. Petersen, at Yale, produced from the late 1990s onward the mechanistic account of muscle insulin resistance in terms of intramyocellular lipid accumulation and mitochondrial dysfunction, using magnetic resonance spectroscopy to see inside human muscle in vivo. Their work established that reduced mitochondrial density in skeletal muscle is a causal contributor to insulin resistance, not merely a correlate.

Benjamin Bikman, of Brigham Young University, has synthesised this literature for a broader audience in Why We Get Sick (2020) and in an extensive video corpus, defending the position that hyperinsulinemia is not the consequence of insulin resistance but its co-driver, and that both can and should be measured directly rather than inferred from downstream markers.

Robert H. Lustig, of UCSF, has developed the case for fructose-driven hepatic insulin resistance and, since the 2010s, has become one of the most persistent public voices arguing that metabolic syndrome is a mislabel for hyperinsulinemia and that measurement should follow suit.

Ivor Cummins, an engineer turned public science communicator, has popularised the triglyceride-to-HDL ratio as the most accessible proxy for insulin resistance available on a routine lipid panel, and has documented in detail the gap between what the research literature knows and what the ordinary clinical encounter delivers.

Four decades. Five researchers of first rank. A pattern that repeats in every developed and every developing country. And a laboratory panel that still, in 2026, does not contain a fasting insulin as a default line.

What the standard panel shows, and what it hides

Glucose is defended, insulin pays for the defence

Blood glucose is one of the most tightly regulated variables in human physiology. The pancreas, the liver, the adrenal glands, and the sympathetic nervous system cooperate to keep it within a narrow band around 70 to 100 mg/dL fasting. That regulation is not a sign that the metabolic system is healthy. It is a sign that the metabolic system is working, at whatever cost, to keep glucose in range.

When a person’s peripheral tissues become resistant to insulin, glucose disposal into skeletal muscle is impaired, and in the adipose tissue insulin fails to restrain lipolysis, releasing free fatty acids that in turn feed back on muscle resistance. The pancreatic beta cells respond by secreting more insulin. If the compensation is adequate, blood glucose stays normal. The person is, at that point, profoundly insulin-resistant with a normal fasting glucose. This state can persist for years, even decades. It is the state Kraft’s patterns II, III, and IV describe. It is invisible to any test that reads only glucose.

Glucose, in this state, is what one might call tightly (dys)regulated: held in range by an increasingly desperate insulin secretion whose cost is written on every other organ. To read glucose alone and pronounce the metabolism healthy is to read the fuel gauge of a burning engine.

The panel is calibrated to see the fire, not the smoke

The routine laboratory panel offered in ordinary primary care includes fasting glucose, sometimes HbA1c, occasionally a lipid panel that reports total cholesterol, LDL, HDL, and triglycerides. It does not, by default, include fasting insulin, a HOMA-IR calculation, C-peptide, a post-load insulin, or any Kraft-style dynamic assessment.

Each of these omissions has a specific consequence:

• No fasting insulin means the compensatory hyperinsulinemia that precedes diabetes by ten to twenty years is not seen.
• No HOMA-IR means the standard research proxy for insulin resistance, calculable from fasting glucose and fasting insulin with no additional test, is not calculated.
• No post-load insulin means Kraft’s patterns cannot be identified even in a person who is given an OGTT for glucose alone.
• No C-peptide means the distinction between endogenous insulin production and injected insulin cannot be made in ambiguous cases.

The clinical consequence is that insulin resistance is diagnosed, when it is diagnosed at all, only after it has produced its complications. Hypertension, dyslipidemia, and central adiposity are treated as three separate diseases with three separate prescriptions, when they are one disease that a single measurement would name. See Calibrated for Nothing on the general architecture of this failure, and The False Negative of Serum Potassium for the parallel case in which a homeostatically defended serum value hides a systemic derangement.

What triglycerides and HDL can tell in the meantime

For a person whose ordinary panel does not include insulin, the triglyceride-to-HDL ratio provides the most practical available proxy. Elevated triglycerides in the absence of a compensatory rise in HDL cholesterol is one of the more specific lipid signatures of insulin resistance. Ratios above 2.0 (in conventional US units, mg/dL) or 1.0 (in SI units, mmol/L) begin to suggest metabolic dysfunction; ratios above 3.5 or 4 are strongly suggestive. This is not a substitute for a fasting insulin. It is what a person can extract from the panel they are already given.

The syndrome that speaks and the panel that does not listen

Hyperinsulinemia advertises itself. The set of symptoms and signs commonly present, individually or in combination, in a person with clinically significant insulin resistance is well characterised in the literature:

• Central adiposity, particularly a waist circumference disproportionate to overall body mass.
• Hypertension, often resistant to lifestyle intervention that does not address insulin directly.
• Dyslipidemia: elevated triglycerides, low HDL, small dense LDL particles.
• Non-alcoholic fatty liver disease, increasingly the leading cause of chronic liver pathology in the industrialised world.
• Reactive hypoglycemia two to four hours after a carbohydrate-rich meal, with the associated symptoms of tremor, sweating, anxiety, and craving for further carbohydrate.
• Postprandial somnolence, the strong desire to sleep within an hour of eating.
• Cognitive symptoms: difficulty concentrating, mental fog, memory lapses.
• Skin markers: acanthosis nigricans (dark velvety patches at the nape of the neck, axillae, or groin), skin tags, and adult acne.
• Polycystic ovary syndrome in women, with the associated hormonal and reproductive consequences.
• Nocturnal awakenings, particularly in the 2 to 4 a.m. window, often attributable to counter-regulatory cortisol and adrenaline release.

A person can present with several of these, be seen in successive appointments over years, receive an antihypertensive, a statin, and general dietary advice, and never have their insulin measured. The diagnosis is not hidden. It is out of panel.

Navigating without a compass

Even when the diagnosis is made, or strongly suspected, the absence of routine insulin measurement produces a second, distinct problem: the interventions cannot be calibrated.

Consider a person who reduces carbohydrate intake to address suspected insulin resistance. Some fraction of the population responds robustly to a moderate reduction (say, 100 g of carbohydrate per day) with meaningful weight loss and symptomatic improvement. Another fraction requires substantially deeper restriction, or the addition of intermittent fasting, or exercise interventions, before insulin secretion falls enough to unlock adipose tissue. A third fraction may respond only after months, as intramyocellular lipid is metabolised and mitochondrial function partially restored.

Without insulin measurement, none of these three groups can distinguish itself. A person who plateaus after modest initial weight loss cannot know whether the intervention needs to be intensified, whether a longer time horizon is required, or whether an entirely different lever should be pulled. The instrument that would tell them is the fasting insulin, the HOMA-IR, or the post-meal insulin dynamic. They do not have it.

The clinical consequence, played out at population scale, is a systematic misattribution of intervention failure to willpower. A person who abandons a low-carbohydrate regimen at a plateau is not evidencing a lack of discipline. They are responding rationally to an instrument that gives no signal. The compass is missing. See The Invisible Medical Emergency on the parallel structural neglect of chronic subclinical derangements.

The invisible evidence base

The absence of routine insulin measurement produces a third consequence, structurally more severe than the first two. It systematically distorts the evidence base for insulin-sensitising interventions.

Because glucose is defended by insulin, an intervention that lowers insulin without moving glucose looks, under standard measurement, like a null result. An intervention that improves subjective wellbeing without moving weight or fasting glucose is classified as placebo. An intervention that partially restores mitochondrial function in skeletal muscle produces no signal on any variable measured in a routine trial.

The population of insulin-sensitising interventions is therefore systematically under-recognised, not because these interventions fail, but because they operate on the variable that is not measured. This is a claim about the epistemology of nutritional and lifestyle research, not about any specific substance. The tools of measurement shape the evidence base. Substances and practices that would show effect on insulin but not on glucose have been sifted out of view by the simple fact that no one looked at the right place.

Interventions of mechanistic interest that would benefit from direct insulin measurement include: sustained caloric or carbohydrate restriction; time-restricted eating; resistance and high-intensity exercise; adequate magnesium status (see Right to Adequate Magnesium Levels); thiamine and its derivatives, particularly benfotiamine (see Right to Adequate Thiamine (B1) Status); alpha-lipoic acid; berberine; inositol (myo- and D-chiro-); mitochondrial cofactors including CoQ10 (see Right to Adequate Coenzyme Q10 Status); vitamin D repletion; adequate sleep architecture. The claim is not that each of these is proven to sensitise insulin in every individual. The claim is that the population-scale test that would distinguish which of them works, for whom, and at what dose, cannot be run under a regime that does not measure insulin.

This is a critical parallel of Behind Every Test, an Industry, read in reverse. Where an industry exists behind a test, the test enters routine practice. Where no industry exists behind a measurement, even a mature, cheap, mechanistically central measurement remains out of routine, and the interventions that would show their value on that measurement remain in the shadow.

The measurement exists, and the obstacle is not technical

Insulin is measured by enzyme-linked immunosorbent assay (ELISA), a mature laboratory technique available in essentially every clinical biochemistry laboratory in the world. The reagent cost per assay, at wholesale, is on the order of one dollar. The analytical procedure is standardised, quality-controlled by international reference materials, and validated for use in venous serum.

Dried blood spot (DBS) collection, in use since the 1960s for newborn screening (the Guthrie test for phenylketonuria, congenital hypothyroidism, and other conditions), has been validated for insulin in multiple research settings. Correlation with venous serum insulin is workable, with the standard caveats on hemolysis and stability that any DBS analyte carries. Capillary micro-sampling devices (Tasso, Mitra, and their generation) are already in use in telemedicine for a range of analytes and could be adapted to insulin with no fundamental technical innovation required.

A ten-sample home-collected insulin panel, mailed to a central laboratory, is realisable with the technology of the 1990s. It is not offered as a routine consumer product in essentially any jurisdiction. The blockage is not physical. It is:

1. Prescription gating. In most jurisdictions, an insulin assay requires a physician order, including at direct-to-consumer laboratories that offer other panels without a script.
2. The absence of a commercial actor. The pharmaceutical and device industries built around insulin exist for the treatment of established diabetes, not for the detection of pre-diabetic hyperinsulinemia. There is no drug tied to a fasting insulin measurement in the general population, no glucose meter analogue for insulin, no lobbying voice that pushes for the test to be routine.
3. The absence of a continuous sensor. Unlike glucose, insulin has no wearable continuous sensor. Research toward one has been active for over fifteen years without a product. This is a genuine technological limitation, but it is a limitation on continuous monitoring, not on intermittent measurement.

The measurement that would change the clinical picture is not the one that requires the future. It is the one that already exists and has never been extended to the population that needs it.

The CGM precedent

Continuous glucose monitoring provides the operative precedent. Developed in research settings in the 1990s, prescribed for insulin-dependent diabetes from the 2000s, gated behind a physician order in most jurisdictions through the early 2020s, and released over-the-counter in the United States in 2024 under the Stelo (Dexcom) and Lingo (Abbott) product lines, the CGM traces the arc of a metabolic measurement moving from research through prescription to direct availability.

The public health case for OTC CGM was clear once someone made it: the tool improves the metabolic self-knowledge of any user, not only the insulin-dependent patient. The regulatory shift acknowledged that the paternalism of prescription gating had held back a mature, safe, and clinically valuable measurement from the population that could benefit from it. There is no principled reason for a home-collected fasting insulin panel to sit twenty more years in the same antechamber. The path is traced.

What CGM can and cannot do for the insulin question

Continuous glucose monitoring, now accessible without prescription in some jurisdictions, allows a person to observe their glucose response to real meals in real conditions, over days and weeks. In ecological validity, this is a substantial advance on any single fasting or post-load measurement performed in a laboratory.

For the insulin question, CGM occupies a specific position:

• Superior to standard glucose testing. Density (hundreds of data points per day) and ecological validity (real meals, real activity, real sleep) yield a richer picture of glucose dynamics than a fasting sample or a standardised OGTT.
• Partially informative on insulin. Post-meal glucose excursions of unusual amplitude, prolonged elevations, or reactive hypoglycemia episodes are all indirect signals of insulin dysregulation and can be read as such.
• Structurally silent on the compensatory phase. The Kraft patterns II through IV are defined by hyperinsulinemic responses that still hold glucose in a normal range. CGM, by measuring glucose, cannot see them. A person can wear a CGM for a month, see a well-behaved glucose trace, and remain in an advanced hyperinsulinemic state.

The right formulation is: CGM is a poor person’s field-validity glucose test, better than what routine medicine offers, and not a substitute for direct insulin measurement. Both instruments belong in a metabolic self-knowledge that the current clinical panel refuses to support.

Grounds in international human rights law

The right to health, codified in Article 12 of the International Covenant on Economic, Social and Cultural Rights (ICESCR, 1966), guarantees “the enjoyment of the highest attainable standard of physical and mental health”. The Committee on Economic, Social and Cultural Rights, in General Comment 14 (2000), identified as a core obligation from which no State may derogate, at §43, access to “essential drugs, as from time to time defined under the WHO Action Programme on Essential Drugs”. The General Comment consistently reads health rights as encompassing the “underlying determinants” of health, including “access to health-related education and information”.

A laboratory measurement that is inexpensive, technically mature, mechanistically central to the leading pathology of the industrialised world, and withheld from ordinary clinical use is a paradigmatic case of the gap between the right on paper and the entitlement in practice. See The Instruments Already Exist for the fuller reading of GC14 and the antiretroviral precedent, and A Litigation Brief for the standing and remedies analysis.

The 1978 Alma-Ata Declaration names, among the components of primary health care, “appropriate diagnostic and treatment of common diseases and injuries”. Type 2 diabetes and its precursor hyperinsulinemic states are the definitional common diseases of the 21st century. A primary care system that cannot detect the upstream state of the leading modern pathology has not delivered on Alma-Ata’s promise.

The right to know

The physiological right at issue is neither exotic nor difficult to state. Where a hormone governs a pathology of civilisational scale; where its measurement is inexpensive, technically mature, and clinically validated; where its exclusion from routine practice cannot be justified by cost, safety, or scientific uncertainty; where the population that would benefit from measurement includes essentially every adult in the industrialised world; there is a right, grounded in the existing instruments of the right to health, to have the measurement offered.

That right implies, in the ordinary case, the inclusion of fasting insulin, HOMA-IR, and where appropriate a post-load insulin dynamic in the routine primary-care panel. It implies, in the direct-to-consumer case, the availability of a low-cost home-collected insulin panel comparable to what is now emerging for continuous glucose. It implies, in the research case, that trials of nutritional and lifestyle interventions of plausible insulin-sensitising mechanism should routinely include insulin as an outcome, so that the invisible evidence base becomes visible.

Naming a pathology whose central variable is not measured is possible in medicine. It is not sustainable in a rights framework. The measurement is available. The obstacle is structural. The remedy is administrative and, where administration fails, legal.

Timeline

• 1922 — Banting, Best, Macleod, Collip: isolation of insulin, University of Toronto.
• 1959 — Rosalyn Yalow and Solomon Berson: radioimmunoassay for insulin (Nobel Prize to Yalow, 1977).
• 1975 — Joseph R. Kraft: first publications on the five-pattern OGTT with insulin.
• 1979 — Ralph DeFronzo: hyperinsulinemic-euglycemic clamp technique.
• 1988 — Gerald Reaven: Banting Lecture, introduction of Syndrome X.
• 1988 — Ralph DeFronzo: Banting Lecture, “The Triumvirate”.
• 1997 — First modern ELISA insulin assays widely commercialised for clinical laboratories.
• Late 1990s — Shulman and Petersen at Yale: MR spectroscopy work on intramyocellular lipid and mitochondrial dysfunction.
• 2000 — CESCR General Comment 14 on the right to health.
• 2008 — Joseph R. Kraft: Diabetes Epidemic and You, popular summary of thirty years of data.
• 2020 — Benjamin Bikman: Why We Get Sick, mainstream synthesis.
• 2024 — Continuous glucose monitors released over-the-counter in the United States (Stelo, Lingo).
• 2026 — Fasting insulin remains outside the default primary-care panel in essentially all jurisdictions.

Selected bibliography

• Banting FG, Best CH. The internal secretion of the pancreas. J Lab Clin Med. 1922;7:251–266.
• Kraft JR. Detection of Diabetes Mellitus In Situ (Occult Diabetes). Lab Med. 1975;6(2):10–22.
• DeFronzo RA, Tobin JD, Andres R. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol. 1979;237(3):E214–E223.
• Reaven GM. Banting Lecture 1988. Role of insulin resistance in human disease. Diabetes. 1988;37(12):1595–1607.
• DeFronzo RA. Lilly Lecture 1987. The triumvirate: β-cell, muscle, liver. A collusion responsible for NIDDM. Diabetes. 1988;37(6):667–687.
• Petersen KF, Befroy D, Dufour S, et al. Mitochondrial dysfunction in the elderly: possible role in insulin resistance. Science. 2003;300(5622):1140–1142.
• Kraft JR. Diabetes Epidemic and You. Trafford Publishing, 2008.
• Crofts C, Zinn C, Wheldon M, Schofield G. Hyperinsulinemia: a unifying theory of chronic disease? Diabesity. 2015;1(4):34–43.
• Bikman B. Why We Get Sick: The Hidden Epidemic at the Root of Most Chronic Disease. BenBella Books, 2020.
• Lustig RH. Metabolical: The Lure and the Lies of Processed Food, Nutrition, and Modern Medicine. Harper Wave, 2021.
• Committee on Economic, Social and Cultural Rights. General Comment No. 14: The Right to the Highest Attainable Standard of Health. E/C.12/2000/4. 2000.
• World Health Organization. Declaration of Alma-Ata, International Conference on Primary Health Care. 1978.