Fasting insulin is one of the most underused markers in preventive metabolic medicine. Many patients are told they are fine because their fasting glucose or HbA1c sits inside the laboratory reference range. But glucose is often the last thing to move. The body works hard to keep blood sugar stable, and it does that by producing more insulin. By the time glucose becomes obviously abnormal, the compensatory process may have been running for years, quietly driving visceral fat accumulation, cardiovascular risk, fatigue, and metabolic decline.
Fasting insulin is not offered on the NHS as part of routine metabolic screening. It is available through private blood testing, and it is a standard component of Veridian's Energy Screen and Core Metabolic Assessment. The reason it matters so much is simple: if glucose looks normal but insulin is already elevated, the system is under strain, and the best time to act is now, while the pattern is still highly reversible.
Clinical bottom line: normal glucose does not always mean normal metabolism. Often it means insulin is still compensating, successfully maintaining glucose at the cost of chronically elevated insulin levels and the downstream consequences that follow.
Why standard glucose markers can miss early dysfunction
Early insulin resistance is fundamentally a story of compensation. Skeletal muscle, which handles roughly 80% of postprandial glucose disposal, becomes less responsive to insulin's signal. The liver becomes less reliable at suppressing glucose output between meals. The pancreas answers by secreting more insulin to maintain normal blood glucose levels. This compensation can be remarkably effective for a surprisingly long time. The patient may already have:
- Central weight gain or waist circumference creeping above 90 cm (men) or 80 cm (women).
- Afternoon energy crashes, brain fog, and carbohydrate cravings.
- Poor recovery from exercise and disrupted sleep quality.
- Elevated triglycerides, low HDL, or rising blood pressure, even while on no medication.
- Fatty liver (MASLD) on an abdominal ultrasound, even in lean-appearing individuals.
And yet their standard blood test (fasting glucose 5.1, HbA1c 36 mmol/mol) shows "nothing to worry about." The compensatory insulin has held the glucose number in the normal range, masking the metabolic stress. Fasting insulin directly reveals whether this compensation is already happening.
What does fasting insulin actually measure, and what is a normal range?
Fasting insulin measures the concentration of insulin in the blood after a minimum of eight hours without eating or drinking anything caloric. It reflects the insulin required to maintain baseline glucose stability in the resting, non-stimulated state.
The laboratory "normal" range in the UK is typically quoted as 2–25 mIU/L or 3–17 mIU/L depending on the assay. These are population-derived reference ranges, not optimal ranges. A fasting insulin of 22 mIU/L may be "within normal" by laboratory standards while representing significant hyperinsulinaemia that is clinically meaningful for cardiovascular and metabolic risk.
In preventive and longevity medicine, the target is substantially lower. Most clinicians working in metabolic health, including those following the Medicine 3.0 frameworks advocated by researchers such as Peter Attia, use the following as a working guide:
- Optimal: fasting insulin below 5–8 mIU/L in a lean, metabolically healthy individual.
- Borderline: fasting insulin 8–12 mIU/L, warrants dietary and lifestyle attention.
- Elevated (hyperinsulinaemia): fasting insulin above 12–15 mIU/L, high priority for intervention.
- Significantly elevated: above 20 mIU/L, usually reflects established insulin resistance; associated with markedly elevated cardiometabolic risk.
These thresholds are not universally agreed upon, and they depend on assay calibration. The important clinical point is that context matters: a fasting insulin of 14 mIU/L in a 58-year-old with a waist-to-height ratio of 0.6, elevated triglycerides, and a family history of diabetes has a different meaning than the same number in a lean 32-year-old who exercises four times a week.
HOMA-IR: a more useful way to interpret the result
HOMA-IR (Homeostatic Model Assessment of Insulin Resistance) is a calculation that combines fasting insulin and fasting glucose to give a better estimate of insulin resistance than either marker alone. The formula is: HOMA-IR = (fasting insulin in mIU/L × fasting glucose in mmol/L) ÷ 22.5.
A HOMA-IR below 1.0 is generally considered insulin-sensitive. Between 1.0 and 1.9 is borderline. Above 2.0 suggests meaningful insulin resistance. Above 2.9 is associated with significant metabolic syndrome risk. Above 5.0 is often seen in established type 2 diabetes.
HOMA-IR is particularly useful because it can identify individuals who have mildly elevated insulin with slightly elevated glucose, both within their respective "normal" ranges, but whose combination is clinically significant. It also tracks improvement well during intervention: a drop in HOMA-IR over twelve weeks is a concrete confirmation that the metabolic pattern is moving in the right direction.
Why elevated insulin matters beyond glucose
Elevated insulin is not just a numerical signal. It actively changes the physiology in multiple directions simultaneously:
Fat metabolism: insulin suppresses lipolysis, the release of fatty acids from adipose tissue for energy. Chronically elevated insulin makes fat-burning physiologically harder, contributing to visceral fat accumulation even in people eating a relatively moderate calorie intake.
Cardiovascular risk: hyperinsulinaemia drives the atherogenic dyslipidaemia cluster, including elevated triglycerides, low HDL, and a shift towards smaller, denser LDL particles (elevated ApoB), all from the liver's response to excess insulin signalling. This is why fasting insulin should rarely be interpreted without also measuring ApoB, triglycerides, and HDL-C.
Blood pressure: insulin promotes renal sodium retention and stimulates the sympathetic nervous system, both of which raise blood pressure. The insulin-hypertension connection helps explain why metabolic syndrome clusters with hypertension so consistently.
Energy and cognition: the energy instability many patients describe, including afternoon crashes, poor concentration, and carbohydrate cravings, maps directly onto the glucose variability that accompanies insulin resistance. Glucose rises sharply after meals, triggers a large insulin response, and falls more steeply than it should. CGM data often confirms this pattern in people who assume their energy symptoms are just "normal."
Who should have a fasting insulin test?
The following groups have the strongest case for fasting insulin as part of a baseline metabolic screen:
- Anyone with a waist circumference above 94 cm (men) or 80 cm (women), regardless of BMI.
- Anyone with triglycerides above 1.5 mmol/L on a standard lipid panel, where elevated TG is often the earliest lipid signal of insulin resistance.
- Women with PCOS, where insulin resistance is present in 65–80% of cases and drives the hormonal dysregulation, yet fasting insulin is not routinely tested in most NHS PCOS workups.
- Anyone with unexplained fatigue, energy instability, cravings, or poor sleep quality alongside any metabolic risk factors.
- Anyone with a family history of type 2 diabetes, cardiovascular disease, or metabolic syndrome.
- Anyone who wants a complete metabolic health picture rather than a screen designed to detect disease after it has declared itself.
What actually improves fasting insulin
- Resistance training: increases GLUT4 transporter density in muscle cells, improving glucose disposal per unit of insulin. Three sessions per week is the minimum effective dose.
- Carbohydrate quality: replacing refined carbohydrates and sugary drinks with fibre-rich whole foods and adequate protein reduces the insulin demand per meal significantly.
- Waist reduction: each kilogram of visceral fat reduction has disproportionate benefit for insulin sensitivity compared to equivalent subcutaneous fat loss.
- Sleep consistency: four nights of sleep below six hours reduces insulin sensitivity by 25% in experimental models; consistently achieving 7–9 hours produces rapid improvements in fasting insulin in sleep-deprived individuals.
- Reduced meal frequency where appropriate: for some individuals, reducing grazing and allowing longer inter-meal intervals lowers overall insulin exposure and improves sensitivity.
- Objective measurement so progress is driven by data rather than guesswork; retesting fasting insulin and HOMA-IR at 8–12 weeks confirms whether the intervention is working at a metabolic level.
The strongest interventions are usually unglamorous but robustly effective: better training stimulus, improved sleep, better food quality, and reduced waist circumference. When indicated, in cases of more advanced insulin resistance or concurrent conditions, medication or more intensive clinical support can also play a role. But the key is spotting the problem early enough that the system is still highly reversible, rather than waiting for glucose to cross a diagnostic threshold.
Related reading: Reversing metabolic syndrome: what the evidence supports · ApoB vs LDL: cardiovascular particle count explained · Lp(a), ApoB, and triglycerides: the triple cardiovascular threat