Fasting Insulin Longevity-Medicine Target Ranges

Fasting insulin is a blood test that measures circulating insulin after an 8 to 12 hour fast. It is distinct from fasting glucose and HbA1c, though it is often interpreted alongside them, and distinct from C-peptide, which measures endogenous insulin production more indirectly. It is not part of a standard basic or comprehensive metabolic panel and has to be ordered separately.

The direct answer: most clinical laboratories treat fasting insulin as abnormal only above roughly 20 to 25 µIU/mL, depending on the assay, while longevity-focused clinicians commonly describe 2 to 6 µIU/mL with a HOMA-IR under 1.0 as an optimization target. That lower target comes from population cohort associations between fasting insulin and later metabolic and cardiovascular outcomes, not from an FDA-approved indication, a diabetes-society treatment goal, or a randomized trial that tested "treat to under 6" against usual care. The American Diabetes Association endorses HOMA-IR as a practical clinical surrogate for insulin resistance but does not publish a specific fasting insulin number for non-diabetic adults to aim for. Because insulin resistance is usually compensated by higher insulin output long before glucose rises, an elevated fasting insulin in someone with a "normal" glucose and A1c can be a genuinely early signal, but exactly which number should trigger treatment is a matter of clinical judgment layered on top of incomplete evidence, not a settled figure.

Why insulin often moves before glucose does

When cells become less responsive to insulin, the pancreas compensates by secreting more of it, which can keep blood glucose in the "normal" range for years even as underlying insulin resistance worsens. By the time fasting glucose crosses into prediabetic territory, that compensatory hyperinsulinemia has usually been present for some time. This sequence is well established in the endocrinology literature and is the core rationale for checking fasting insulin in people whose glucose and A1c look reassuring.

Insulin acts on skeletal muscle, liver, and fat tissue to facilitate glucose uptake, with skeletal muscle responsible for the majority of insulin-stimulated glucose clearance in the postprandial state. Progressive loss of receptor sensitivity may trigger elevated fasting insulin levels, which can further impair receptor function. The speed and degree of this feedback loop differ among individuals, and numerical findings from previously cited studies could not be confirmed against original sources and have therefore been excluded.

Why the standard lab reference range is a weak optimization target

Most commercial labs report fasting insulin as "normal" across a wide range, commonly extending into the high teens or twenties (µIU/mL), because that range is built from the distribution of results in a general patient population, a population that includes a large share of people with some degree of insulin resistance. A range built this way tells you where you fall relative to that population. It does not tell you where insulin sensitivity is biologically lowest-risk. A result in the high teens can sit inside a lab's normal range while still corresponding to meaningful insulin resistance on more sensitive measures like HOMA-IR.

Clinical "normal" versus the longevity-medicine "optimal" target

These are two different questions. "Is this normal" asks whether you fall below the threshold most clinicians use to flag overt insulin resistance, commonly cited as fasting insulin above roughly 10 to 15 µIU/mL or HOMA-IR above roughly 2.5 to 3.0. "Is this optimal" asks whether your insulin sensitivity sits in the range that population cohort studies associate with the lowest long-term cardiometabolic risk, which longevity-focused clinicians generally describe as 2 to 6 µIU/mL with HOMA-IR under 1.0.

The American Diabetes Association's Standards of Care describes HOMA-IR as correlating reasonably well with clamp-based insulin sensitivity measurement and as a practical surrogate for clinical and research use (ADA Standards of Care). That document supports HOMA-IR as a useful calculation; it does not endorse a specific low fasting insulin number as a treatment target for people without diabetes.

The 2 to 6 µIU/mL figure used in longevity medicine is drawn from observational cohort work showing that risk of metabolic and cardiovascular outcomes tends to rise gradually as fasting insulin climbs, even within what labs call the normal range, with people in the lowest observed range showing the lowest associated risk. This is a real and clinically interesting pattern. It is not the same as trial evidence that treating someone down to a specific number changes their outcomes. Cohort associations can be confounded by diet, activity, body composition, and other factors that travel together with low fasting insulin, and the specific point estimates (odds ratios, hazard ratios, exact cohort sizes) attached to individual studies in earlier drafts of this material could not be verified against the cited sources and have been removed rather than restated as fact. A reader who wants those exact figures should ask their clinician to pull the primary study rather than rely on a secondhand number.

HOMA-IR: turning two numbers into one

HOMA-IR converts a fasting insulin and fasting glucose pair into a single index: (fasting insulin in µIU/mL × fasting glucose in mmol/L) ÷ 22.5. It was developed as a simpler alternative to the euglycemic clamp, the gold-standard research method for measuring insulin sensitivity, and multiple validation studies have found it correlates reasonably well with clamp results across different populations, though not perfectly.

A decision framework for what a fasting insulin (and HOMA-IR) result should trigger

Use this reference table as a basis for discussion with your healthcare provider, not as a means of self-assessment. It is not a substitute for professional clinical evaluation and does not establish treatment or dosing recommendations for individual patients.

Fasting insulin in PCOS

Polycystic ovary syndrome is a common endocrine condition in reproductive-age women; the World Health Organization's patient fact sheet describes it as affecting a meaningful share of women of reproductive age globally, with insulin resistance present in a large majority of cases regardless of body weight (WHO, PCOS fact sheet, accessed 2025). Elevated insulin is thought to both result from and worsen the hormonal picture in PCOS, because insulin can stimulate ovarian tissue to overproduce androgens through insulin and IGF-1 receptor signaling.

Because of this mechanism, some clinicians treat a lower fasting insulin threshold in PCOS than they would in someone without PCOS, aiming below roughly 8 µIU/mL rather than the general population's clinical cutoff of 10 to 15. This is a physiologically reasonable extrapolation, not a number specified in current guidelines.

The Endocrine Society's clinical practice guideline on PCOS supports the use of insulin-sensitizing agents, particularly metformin, for women with PCOS who have evidence of insulin resistance or metabolic syndrome and an insufficient response to lifestyle therapy alone. That guideline supports the category of treatment; it does not specify a fasting insulin number as the trigger, and the exact wording attributed to it in earlier drafts of this material could not be verified against a confirmed source and should be checked against the current published guideline before being quoted directly.

Metformin and combined myo-inositol/D-chiro-inositol regimens have both shown reductions in fasting insulin in randomized trials in women with PCOS. The magnitude of those reductions varies across studies, and specific point estimates from individual trials referenced in earlier drafts of this page could not be confirmed against the primary publications and have been removed rather than restated. A clinician weighing metformin against inositol supplementation for a specific patient should review the current trial literature directly rather than rely on secondhand effect sizes.

Fasting insulin and cardiovascular risk: what is established versus plausible

Hyperinsulinemia is generally considered a cardiovascular risk marker that is at least partly independent of obesity and dyslipidemia, acting through mechanisms that include vascular smooth muscle proliferation, reduced nitric oxide availability, and impaired fibrinolysis. These mechanisms are described in cardiovascular physiology literature and are biologically plausible contributors to atherosclerosis at insulin levels below those seen in overt diabetes.

Cohort studies have reported associations between higher fasting insulin and increased risk of cardiovascular events, independent of fasting glucose in some analyses. The specific hazard ratios and confidence intervals attributed to named trials and meta-analyses in earlier versions of this article could not be verified against the underlying publications and have been removed.

The clinical trial evidence on whether lowering insulin itself improves cardiovascular outcomes is limited and mixed. A large trial of exogenous insulin therapy in people with dysglycemia found no significant reduction in cardiovascular outcomes compared with standard care, despite improved glucose control, even though that trial raised circulating insulin levels rather than lowering them. This is consistent with the idea that the route by which insulin is lowered, and what else changes alongside it, matters. Lifestyle interventions that lower endogenous fasting insulin, such as resistance training, lower-glycemic dietary patterns, and weight loss, are more consistently linked to cardiovascular benefit in randomized trials than pharmacologic insulin-lowering alone. This distinction is a genuine boundary condition that a generic reference page on this test typically does not spell out: a lower fasting insulin number achieved through different means is not automatically equivalent in terms of expected benefit.

Testing it accurately

Before the draw: fast 8 to 12 hours, water is permitted, avoid coffee (it affects insulin secretion through more than one mechanism) and any caloric intake. Vigorous exercise the morning of the test can transiently lower insulin and produce a falsely reassuring result, so most clinicians ask patients to rest that morning. If you take a beta-blocker and cannot pause it, note this on the requisition, since these medications can raise baseline insulin in some patients.

Assay variability: insulin immunoassays are known to vary meaningfully across different laboratory platforms, which means a single result from one lab is not always directly comparable to a result from a different lab months or years later. Tracking trends on the same platform over time is more informative than comparing isolated results across labs. Ask your lab which assay it uses, and try to stay with the same lab for serial monitoring.

Pairing with other tests: fasting insulin is most useful alongside fasting glucose, which allows HOMA-IR calculation. Fasting C-peptide can help distinguish endogenous insulin production from exogenous insulin in people already on insulin therapy, since C-peptide is co-secreted with insulin but is cleared more slowly and is a longer-lived surrogate marker.

What moves the number

Factors generally associated with higher fasting insulin include diets high in refined carbohydrate and added sugar, visceral adiposity, insufficient sleep, and prolonged sedentary time, along with certain medications such as glucocorticoids, some antipsychotics, and some beta-blockers.

Factors generally associated with lower fasting insulin include resistance training, lower-glycemic and lower-carbohydrate dietary patterns, weight loss, and in some studies, time-restricted eating windows. The relative size of these effects varies across trials and populations; specific percentage reductions cited in earlier versions of this page for individual studies could not be verified against primary sources and have been removed. Readers who want exact numbers for a specific intervention should look up the primary trial rather than rely on a secondhand figure.

Pharmacologic options when lifestyle change is not enough

When fasting insulin remains elevated after a genuine trial of dietary and activity change (commonly 8 to 16 weeks), clinicians may consider insulin-sensitizing medications. This is a clinical judgment call, not a universally protocolized step, and any medication decision, including dosing, needs to be individualized by a prescribing clinician.

Metformin improves hepatic insulin clearance and reduces hepatic glucose output, and is commonly used off-label for metabolic optimization outside of a diabetes diagnosis. This off-label use is a matter of clinical judgment supported by physiologic rationale and trial data in related populations (particularly PCOS), not an FDA-approved indication for insulin optimization in people without diabetes.

GLP-1 receptor agonists (semaglutide, tirzepatide, liraglutide) are FDA-approved for type 2 diabetes and, at certain doses, for chronic weight management. They lower fasting insulin indirectly, largely through weight loss and reduced visceral fat, which improves insulin sensitivity. Trials such as STEP-1 (semaglutide) and SURMOUNT-1 (tirzepatide) reported substantial weight loss at roughly a year of follow-up, with corresponding improvements in metabolic markers reported as secondary outcomes. Readers who want the exact percentage figures from these trials should verify them against the original NEJM publications rather than a secondhand summary, since specific numbers repeated in earlier drafts of this page could not be independently confirmed here.

Berberine is a plant-derived compound sold as a dietary supplement, not an FDA-approved drug. Some randomized trials have reported reductions in HOMA-IR comparable to metformin in people with insulin resistance or type 2 diabetes, but berberine is not regulated or manufactured with the same quality controls as a prescription drug, interacts with several medications (including some statins and immunosuppressants), and should not be substituted for prescribed therapy without a clinician's input.

A note on public figures and longevity-medicine targets

Some physicians whose practice focuses on longevity medicine, most visibly Peter Attia, MD, have publicly described targeting fasting insulin below roughly 6 µIU/mL and HOMA-IR below 1.0 in their patients, citing the continuous-risk pattern seen in population cohorts rather than a binary clinical cutoff. This reflects one clinician's stated practice approach and a broader trend in longevity-focused primary care. It is not a peer-reviewed guideline, and the specific wording of any public statement attributed to a named individual should be verified against a primary source (an interview, article, or podcast transcript) before being quoted directly in a published piece. This section describes a documented practice pattern, not clinical evidence in itself.

What is established, what is plausible, and what is not

Established: insulin resistance is often compensated by rising insulin output long before fasting glucose becomes abnormal, so fasting insulin can flag early metabolic change. HOMA-IR is a reasonably validated surrogate for insulin sensitivity, endorsed by the ADA as a practical clinical tool. PCOS is strongly associated with insulin resistance in most affected women. Lifestyle interventions (resistance training, lower-glycemic diets, weight loss) reliably lower fasting insulin in trials.

Plausible but not proven by trial evidence: that treating fasting insulin down to a specific number such as under 6 µIU/mL, as opposed to simply staying under commonly used clinical thresholds, produces measurably better long-term health outcomes for an individual patient. That pharmacologically lowering insulin, independent of the weight loss or lifestyle change that usually accompanies it, directly reduces cardiovascular or mortality risk.

Not established: a single universal fasting insulin or HOMA-IR cutoff that applies across all assays, labs, and populations. Fasting insulin also is not, by itself, a diagnostic test for diabetes or PCOS; it is one input into a broader clinical picture that includes glucose, A1c, lipids, blood pressure, and in PCOS, hormonal and menstrual history.

When to seek care rather than self-manage this number

A single elevated fasting insulin result is not an emergency. However, new or worsening symptoms such as unexplained rapid weight change, persistent fatigue with high thirst and urination, irregular or absent periods with signs of androgen excess, or any symptoms suggestive of diabetic ketoacidosis (nausea, vomiting, abdominal pain, confusion) warrant prompt medical evaluation rather than waiting on a scheduled recheck.