HOMA-IR & Insulin Resistance: The Evidence
Why a score above 1.9 matters, how it is measured, and the many ways insulin resistance reveals itself across the body
Scepticism about insulin resistance being a "real" or clinically meaningful condition is understandable — it is not yet a standard NHS test, and many people have never heard of HOMA-IR. But the evidence base is large, old and remarkably consistent. Insulin resistance is not a theory: it is a measurable state reflected in dozens of independent biomarkers simultaneously shifting in the same direction. Below is the evidence — from the origins of the HOMA-IR score to the full spectrum of what goes wrong in the body when insulin signalling is impaired.
In Summary: The Case for Taking HOMA-IR Above 1.9 Seriously
HOMA-IR above 1.9 is not an arbitrary line. It marks the zone where the full cluster of metabolic markers — lipids, liver enzymes, blood pressure, hormones, inflammatory markers, uric acid and physical signs — begins to shift in a consistently unfavourable direction. The gold-standard hyperinsulinaemic-euglycaemic clamp validates HOMA-IR as an accurate reflection of what is happening at the cellular level. And the wide, independent convergence of biomarkers across entirely different systems — all pointing to the same underlying disruption — is what transforms insulin resistance from a theoretical concept into one of the most evidence-based constructs in modern metabolic medicine.
The solution — reducing dietary carbohydrates and restoring periods of not eating — addresses the root cause directly: insulin levels fall, cells recover their sensitivity, autophagy resumes, and the entire cluster of markers begins to normalise.
What Is HOMA-IR and Where Does the 1.9 Threshold Come From?
HOMA-IR stands for Homeostatic Model Assessment of Insulin Resistance. It was developed at Oxford University in 1985 by Matthews, Hosker, Rudenski and colleagues, and published in Diabetologia. The formula is simple:
HOMA-IR = (Fasting Insulin in µU/mL × Fasting Glucose in mmol/L) ÷ 22.5
or, if glucose is in mg/dL: (Fasting Insulin × Fasting Glucose) ÷ 405
The score estimates how hard the pancreas must work to keep blood glucose stable. The higher the score, the more insulin resistance is present. See key studies below.
Why 1.9?
There is no single universal threshold for HOMA-IR — this is one source of confusion. Different populations, ages and ethnicities show different optimal cut-offs. However, a consistent picture emerges across the literature:
| HOMA-IR Score | Interpretation (Functional Medicine Standard) | Clinical / Research Context |
|---|---|---|
| Below 1.0 | Optimal insulin sensitivity | Goal for metabolic health |
| 1.0 – 1.9 | Borderline — early impairment beginning | Functional medicine practitioners intervene here |
| 2.0 – 2.9 | Insulin resistance present | Consistent with most peer-reviewed cut-offs; NHANES uses ≥2.5 for US adults |
| 3.0 and above | Significant insulin resistance | Associated with metabolic syndrome and type 2 diabetes risk |
The functional medicine threshold of 1.9 (marking borderline) reflects the evidence that metabolic risk begins rising well before the conventional 2.0–2.5 clinical cut-offs. The landmark 2013 EPIRCE cross-sectional study by Gayoso-Diz et al., involving 2,459 Spanish adults, found that HOMA-IR cut-offs for identifying those at elevated cardiometabolic risk vary by age and gender — and that basing thresholds on population percentiles (rather than actual metabolic risk) misses early cases. Their approach of using cardiometabolic risk markers as the reference — rather than just statistical norms — is why more sensitive thresholds (1.5–1.9) are now used in clinical practice.
Chinese and South Asian populations tend to develop metabolic complications at lower HOMA-IR scores (1.4–1.78), while US population data (NHANES III) found a median HOMA-IR of 2.2 in adults without diabetes — suggesting a significant proportion of the population already has impaired insulin sensitivity even by conventional standards.
The Sceptic's Question: How Do We Know Insulin Resistance Is Real?
The gold standard for measuring insulin resistance is the hyperinsulinaemic-euglycaemic clamp — a controlled infusion of insulin with glucose measured in real time to assess how efficiently cells respond. HOMA-IR was validated against this gold standard and correlates strongly with it. But more importantly: insulin resistance does not exist in isolation. When HOMA-IR rises, a predictable constellation of other independent biomarkers moves in concert — triglycerides rise, HDL falls, blood pressure climbs, liver enzymes elevate, uric acid increases, inflammatory markers go up, and sex hormone balance shifts. These are not opinions — they are objective measurements. The consistency with which these markers co-vary is what makes insulin resistance one of the most robustly evidenced constructs in metabolic medicine.
What Else Goes Wrong: The Full Biomarker Picture
Insulin resistance is not a single-tissue problem. Insulin receptors are present throughout the body, and when cells stop responding effectively, the downstream effects are measurable across multiple systems simultaneously. Below are the key categories of measurable change.
Blood Sugar Markers
Fasting Glucose
Elevated fasting glucose reflects hepatic insulin resistance — the liver continues releasing glucose overnight when it should not. Persistent elevations signal that the liver is no longer responding properly to insulin's "stop producing glucose" signal.
HbA1c (Glycated Haemoglobin)
Reflects average blood glucose over the previous 3 months. When HbA1c is still normal but HOMA-IR is elevated, this confirms the pre-diabetic metabolic stage — the pancreas is still compensating and glucose has not yet entered the dysglycaemic range. It is a later-stage signal than HOMA-IR.
Fasting Insulin
Often the earliest signal. When cells resist insulin, the pancreas compensates by producing more of it. Chronically elevated fasting insulin is itself a driver of further insulin resistance — creating a reinforcing cycle that predates any glucose abnormality.
C-Peptide
Produced in equal amounts to insulin, C-peptide provides a direct measure of how much insulin the pancreas is manufacturing. High C-peptide in the absence of elevated glucose confirms compensatory hyperinsulinaemia — the pancreas working overtime.
Lipid Markers
Triglycerides (Elevated)
One of the most consistent markers of insulin resistance. Insulin normally suppresses triglyceride production in the liver; when resistance develops, the liver continues overproducing VLDL (very low-density lipoprotein), raising triglycerides in the blood. Elevated triglycerides (≥1.7 mmol/L or ≥150 mg/dL) are a formal criterion for metabolic syndrome.
HDL Cholesterol (Reduced)
HDL falls as triglycerides rise — they are metabolically linked through cholesteryl ester transfer protein (CETP). Low HDL (below 1.0 mmol/L in men, 1.3 mmol/L in women) is a formal metabolic syndrome criterion. The TG:HDL ratio is now one of the most validated surrogate markers for insulin resistance, with an AUC of up to 0.88 in some populations.
TyG Index (Triglyceride-Glucose)
Calculated from triglycerides and fasting glucose, the TyG index is recognised as the leading alternative surrogate for insulin resistance when fasting insulin is unavailable. It correlates strongly with the hyperinsulinaemic-euglycaemic clamp — the gold standard. A TyG above approximately 8.5 indicates insulin resistance in most Western populations.
Small Dense LDL Particles (sdLDL)
Insulin resistance shifts LDL from large, buoyant particles to small, dense ones. Small dense LDL particles are more atherogenic — they penetrate arterial walls more easily and are more prone to oxidation. This is why standard LDL cholesterol may appear normal while actual cardiovascular risk is elevated.
Liver Markers
ALT (Alanine Aminotransferase)
ALT rises as fat accumulates in the liver — a condition now called Metabolic dysfunction-Associated Steatotic Liver Disease (MASLD, formerly NAFLD). Insulin resistance is the primary driver of hepatic fat accumulation. Research has confirmed HOMA-IR as an independent determinant of elevated ALT, AST and GGT even after adjusting for BMI, age and cholesterol.
GGT (Gamma-Glutamyl Transferase)
GGT is a sensitive early indicator of liver metabolic stress and correlates strongly with insulin resistance. It often rises before ALT does, making it a useful early warning signal. Some research suggests GGT may even be a stronger predictor of future type 2 diabetes than ALT alone.
AST (Aspartate Aminotransferase)
Like ALT, AST is elevated in liver injury from fat infiltration associated with insulin resistance. The AST:ALT ratio provides additional information — a ratio below 1.0 suggests non-alcoholic fatty liver, consistent with metabolic disease rather than alcohol-related damage.
Inflammation Markers
hs-CRP (High-Sensitivity C-Reactive Protein)
A marker of systemic low-grade inflammation, hs-CRP consistently co-travels with insulin resistance and metabolic syndrome. Adipose tissue — particularly visceral fat — secretes inflammatory cytokines (TNF-α, IL-6) that both cause and are worsened by insulin resistance, creating a bidirectional inflammatory loop.
Uric Acid (Elevated)
Insulin resistance reduces the kidneys' ability to excrete uric acid by increasing sodium reabsorption, leading to hyperuricaemia. A statistically significant correlation between uric acid levels and HOMA-IR has been demonstrated repeatedly (r = 0.299, p <0.001 in multiple studies). High uric acid is associated with hypertension, gout and cardiovascular disease — all conditions in the insulin resistance cluster.
Blood Pressure (Elevated)
Insulin normally promotes vasodilation via nitric oxide. In insulin-resistant states, this vasodilatory effect is lost while insulin's sodium-retaining effect in the kidney persists — raising blood pressure. Hypertension is a formal metabolic syndrome criterion and tracks closely with HOMA-IR in population studies.
Hormone Markers
SHBG (Sex Hormone-Binding Globulin) — Reduced
Chronically elevated insulin suppresses liver production of SHBG. Low SHBG is one of the earliest and most sensitive markers of insulin resistance — often falling before blood sugar shows any abnormality. It is also a primary driver of the hormonal disruptions seen in PCOS.
Free Testosterone (Elevated in Women)
When SHBG falls, free androgens rise. In women, this produces the hallmark signs of PCOS: irregular cycles, acne, hirsutism, and fertility difficulties. High testosterone combined with low SHBG and insulin resistance form a well-established triad in women's metabolic health.
Free Testosterone (Reduced in Men)
In men, insulin resistance is associated with reduced testosterone production. Visceral fat converts testosterone to oestrogen via aromatase, while the same inflammatory cytokines that drive insulin resistance suppress Leydig cell function in the testes. Low testosterone in men further worsens insulin sensitivity — another reinforcing cycle.
Cortisol (Dysregulated)
Chronic stress elevates cortisol, which directly antagonises insulin action — raising blood glucose and promoting visceral fat storage. This is why chronic psychological stress is both a cause and a consequence of insulin resistance. Sleep deprivation has the same effect: even one week of poor sleep measurably elevates HOMA-IR.
Physical and Skin Signs
Acanthosis Nigricans
Dark, velvety thickening of the skin in folds — especially the back of the neck, armpits and groin — is a direct physical manifestation of insulin excess. Chronically elevated insulin activates keratinocyte IGF-1 receptors, causing skin cell proliferation. It is visible confirmation of what the blood tests show.
Skin Tags (Acrochordons)
Small benign skin growths in skin folds develop through the same IGF-1 receptor mechanism as acanthosis nigricans. Multiple skin tags, particularly on the neck and underarms, are a recognised clinical indicator of insulin excess and are associated with metabolic syndrome.
Central Adiposity (Waist Circumference)
Insulin promotes fat storage, particularly in visceral (intra-abdominal) depots. Visceral fat is metabolically active — it secretes pro-inflammatory cytokines and free fatty acids that directly impair insulin signalling. Waist circumference above 94cm in men or 80cm in women is a metabolic syndrome criterion and strongly correlates with HOMA-IR.
The Convergence Argument: Why This Matters for the Sceptic
For a sceptic, the most compelling evidence is not any single marker — it is the convergence. When insulin resistance is present, the following occur simultaneously and consistently across populations, ages, ethnicities and genders:
Triglycerides rise. HDL falls. Blood pressure increases. Liver enzymes elevate. Uric acid climbs. SHBG drops. hs-CRP increases. Fasting insulin rises. Waist circumference expands. Small dense LDL particles proliferate. Skin tags and acanthosis nigricans appear. — These are not opinions or theories. They are independent laboratory measurements that consistently co-vary with HOMA-IR across thousands of peer-reviewed studies.
None of these markers "know" about each other. They are measured by entirely different laboratory assays, in different tissues, by different mechanisms. Yet they move together — because they share a single upstream cause: the loss of appropriate insulin signalling. This convergence is the most powerful evidence that insulin resistance is a real, measurable, clinically significant state.
The question of whether to use a HOMA-IR threshold of 1.9, 2.0 or 2.5 is a calibration question — different populations need slightly different cut-offs. But the underlying biology is not in question. A HOMA-IR above 1.9 is a signal to look further at the full picture, not a diagnosis in isolation.
Key Studies & Research References
- Homeostasis Model Assessment: Insulin Resistance and Beta-Cell Function from Fasting Plasma Glucose and Insulin Concentrations in Man https://pubmed.ncbi.nlm.nih.gov/3899825/ The original Oxford paper developing the HOMA formula, validated against direct measures of insulin secretion and resistance. Developed at Oxford in 1985, this remains one of the most widely cited papers in metabolic medicine and the foundation for all subsequent HOMA-IR research.
- Insulin Resistance (HOMA-IR) Cut-off Values and the Metabolic Syndrome in a General Adult Population: Effect of Gender and Age — EPIRCE Cross-Sectional Study https://bmcendocrdisord.biomedcentral.com/articles/10.1186/1472-6823-13-47 A large Spanish population study (2,459 adults) establishing that HOMA-IR cut-offs should be based on cardiometabolic risk rather than population percentiles. Demonstrated the range of 1.5–3.0 for clinically meaningful insulin resistance and provided the evidence base for gender- and age-adjusted thresholds now widely used in research.
- The Triglyceride/HDL Ratio as a Surrogate Biomarker for Insulin Resistance https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11274455/ A systematic review confirming the TG:HDL ratio as a practical, cost-effective surrogate marker for insulin resistance across diverse ethnicities. Found average optimal cut-offs of 2.53 for women and 2.8 for men, with strong correlation to HOMA-IR across adults and children.
- Obesity, Insulin Resistance and Their Interaction on Liver Enzymes https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8059853/ Confirmed that insulin resistance is an independent determinant of elevated ALT, AST and GGT in non-diabetic Chinese adults, even after adjusting for age, sex, BMI, triglycerides and cholesterol. This is important evidence that the liver effects of insulin resistance are not simply a consequence of obesity.
- What Is the Relationship Between Serum Uric Acid Level and Insulin Resistance? A Case-Control Study https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10754590/ In 2,530 patients, a statistically significant positive correlation was found between uric acid and HOMA-IR (r = 0.299, p <0.001), and between uric acid and the TG:HDL ratio. Insulin resistance reduces renal urate excretion through increased sodium reabsorption — a direct mechanistic link.
- Skin Disease Related to Metabolic Syndrome in Women https://www.sciencedirect.com/science/article/pii/S2352647519300838 Confirmed that sex hormone-mediated cutaneous conditions including acanthosis nigricans, PCOS, acne vulgaris and pattern alopecia are associated with insulin resistance and elevated metabolic syndrome risk. High testosterone, low SHBG and low oestrogen all increase risk of both metabolic syndrome and type 2 diabetes in women.
- Acanthosis Nigricans — StatPearls (NCBI Bookshelf) https://www.ncbi.nlm.nih.gov/books/NBK431057/ Established that acanthosis nigricans develops because chronically elevated insulin activates keratinocyte IGF-1 receptors, causing skin cell and fibroblast proliferation. The condition is a visible, physical sign of insulin excess — confirming the systemic nature of insulin resistance at the skin level.
- Insulin Resistance — StatPearls (NCBI Bookshelf) https://www.ncbi.nlm.nih.gov/books/NBK507839/ A comprehensive clinical overview confirming that all tissues with insulin receptors can become insulin resistant, with the liver, skeletal muscle and adipose tissue as primary drivers. Formally lists elevated triglycerides (≥150 mg/dL), reduced HDL, elevated blood pressure and elevated fasting glucose as co-occurring criteria in the joint scientific statement on metabolic syndrome.