Insulin resistance: what it means, how it is assessed, and what one test cannot prove
Insulin resistance is a biological continuum in which tissues respond less effectively to insulin. It matters—but routine clinical care has no single universally accepted fasting-insulin or HOMA-IR cutoff that diagnoses it in every adult.
The quick answer
Diagnose established dysglycemia with validated glucose tests; assess insulin resistance as a clinical and metabolic pattern.
Fasting plasma glucose, A1C, and the oral glucose tolerance test have established criteria for prediabetes and diabetes. Fasting insulin and HOMA-IR can estimate insulin resistance in research and selected clinical contexts, but assay variability and population-specific distributions prevent one universal diagnostic threshold. Risk assessment should also examine adiposity and waist, blood pressure, lipids, liver, medicines, sleep, physical activity, family history, and trends.
A continuum—not a switch
Sensitivity varies across tissues and over time. A person does not become universally ‘resistant’ at one magic number.
Glucose tests diagnose dysglycemia
Fasting glucose, A1C, and oral glucose tolerance have guideline criteria; fasting insulin does not replace them.
Treat outcomes, not a surrogate alone
The goal is lower cardiometabolic risk and prevention or management of established disease—not simply a lower HOMA-IR.
The central distinctionInsulin resistance is real physiology. A commercially promoted cutoff is not automatically a validated clinical diagnosis.
Side by side
Common measures answer different questions.
Pattern recognition
Four patterns that deserve different interpretations.
None is a stand-alone diagnosis. The purpose is to decide what established disease, risk, or cause should be evaluated next.
Risk may precede dysglycemia
Central adiposity, family history, gestational diabetes, PCOS, MASLD, medicines, and sleep apnea can justify guideline-based screening and prevention.
Established high-risk state
Confirm appropriate criteria, assess cardiovascular risk, and use structured lifestyle intervention; selected people may need medication discussion.
A surrogate signal
Check assay, fasting conditions, repeatability, phenotype, glucose tests, and whether the result changes a validated decision.
Use diagnostic pathways
Confirm when required, classify the condition, assess symptoms and complications, and initiate evidence-based management—not a supplement-only ‘reversal’ plan.
MERHI ONE Evidence Map
Established clinical uses versus attractive shortcuts.
The rating applies to each exact claim, not to ‘metabolic health’ as a marketing category.
Fasting glucose, A1C, and oral glucose tolerance have validated criteria for diagnosing prediabetes and diabetes.1,3,4
Current U.S. and Brazilian diabetes guidelines use these glycemic tests, with confirmation rules and condition-specific caveats.
The hyperinsulinemic-euglycemic clamp is a reference method for measuring insulin sensitivity.6
It directly evaluates glucose disposal under controlled insulin exposure but is resource intensive and mainly a research tool.
Fasting insulin and derived indices can identify higher-risk patterns at the population level.6,7
They correlate with insulin sensitivity and future dysglycemia, but performance depends on assay, population, age, adiposity, and model.
One HOMA-IR cutoff diagnoses insulin resistance in every adult.1,3,6,7
No globally standardized fasting-insulin assay, reference population, or outcome-linked threshold supports a single universal border.
Lifestyle intervention can reduce progression from prediabetes to type 2 diabetes.2,4
Structured nutrition, activity, and weight-management programs have clinically meaningful prevention evidence; selected high-risk adults may benefit from medication.
Lowering fasting insulin with a supplement proves lower cardiovascular or diabetes risk.1,2
Changing a surrogate does not establish benefit on disease incidence, complications, or mortality.
Two native pathways
Screening language differs; core diagnostic tests agree.
United States — ADA / USPSTF
- ADA diagnostic criteria center on fasting plasma glucose, A1C, random glucose with classic symptoms, and 2-hour oral glucose testing.
- USPSTF recommends screening adults 35–70 with overweight or obesity and offering or referring those with prediabetes to effective preventive interventions.
- Fasting insulin and HOMA-IR are not universal diagnostic criteria for prediabetes or diabetes.
Brazil — Sociedade Brasileira de Diabetes
- The SBD diagnostic pathway likewise uses glycemia, HbA1c, and oral glucose testing with confirmation and clinical-context rules.
- Brazilian screening recommendations and feasibility should be applied using current SBD guidance and individual risk.
- Insulin and HOMA-IR may provide selected context but do not replace established criteria.
What a useful assessment includes
Look for established disease, overall risk, and modifiable drivers.
Validated glycemic testing
Choose fasting glucose, A1C, or oral glucose tolerance according to the question and recognize conditions that distort each test.
Anthropometry and trajectory
Weight history and waist can add risk information without defining health or worth.
Blood pressure and lipoproteins
Cardiovascular risk may be high even when glucose is normal.
Liver and kidney context
MASLD, albuminuria, and kidney function alter both risk and management.
Medicines and endocrine conditions
Glucocorticoids, antipsychotics, PCOS, sleep apnea, and other contexts can contribute.
Behavior and feasibility
Sleep, activity, nutrition, alcohol, stressors, access, and preferences shape a realistic plan.
Common interpretation errors
Metabolic marketing often turns a continuum into a label.
The safest pathway keeps diagnostic criteria, surrogate markers, and treatment outcomes separate.
‘HOMA-IR above X proves disease’
Cutoffs are population and assay dependent.
‘Normal A1C means no metabolic risk’
Lipids, blood pressure, liver, kidney, waist, and post-meal glucose may add risk.
‘High insulin explains every symptom’
Fatigue, hunger, weight change, and skin findings have broad differentials.
‘A CGM diagnoses insulin resistance’
Continuous glucose monitors can show glucose patterns but do not directly measure insulin sensitivity.
‘One supplement reverses insulin resistance’
Outcome evidence, dose, population, safety, and comparator must be specified.
‘Thin people cannot be insulin resistant’
Body size alone neither proves nor excludes altered insulin sensitivity or dysglycemia.
Possible consequences of overdiagnosis
- Anxiety from an unvalidated cutoff
- Unnecessary repeat panels
- Restrictive diets without indication
- Supplement interactions and cost
- Missed diabetes or cardiovascular risk
- Focus on a surrogate instead of outcomes
A practical next step
Build a decision, not a label.
- 01
Define whether the question is diabetes screening, symptom evaluation, risk estimation, or treatment monitoring.
- 02
Use validated glucose tests and repeat or confirm according to the relevant guideline.
- 03
Assess blood pressure, lipoproteins, waist or body trajectory, liver, kidney, medicines, sleep, activity, and family history as appropriate.
- 04
If insulin or HOMA-IR is used, document the assay, context, uncertainty, and decision it could change.
- 05
Track validated outcomes and sustainable behaviors rather than chasing one surrogate number.
Frequently asked questions
Direct answers to practical questions.
Is insulin resistance a real condition?+
It is real physiology and part of several metabolic disorders, but routine care lacks one universal standalone diagnostic test.
What is a normal fasting insulin?+
Laboratory intervals and assays vary; no single universal clinical target applies to every adult.
What HOMA-IR value diagnoses insulin resistance?+
There is no globally validated universal cutoff. Published thresholds are population and method specific.
Can glucose be normal while insulin is high?+
Yes, compensation can preserve glucose, but one high insulin result still requires context and does not prove a complete diagnosis.
Does a CGM measure insulin resistance?+
No. It measures interstitial glucose, not insulin action.
Can insulin resistance improve?+
Yes. Depending on cause and stage, activity, nutrition, weight management, sleep and condition-specific treatment can improve risk and glycemic outcomes.
Scientific sources
The diagnostic limits are part of the evidence.
We prioritize current diabetes guidelines, public-health screening recommendations, official education, and validated physiology methods.
- 01Open source ↗
American Diabetes Association · 2026
Standards of Care in Diabetes — Diagnosis and Classification
- 02Open source ↗
American Diabetes Association · 2026
Prevention or Delay of Diabetes and Associated Comorbidities
- 03Open source ↗
Sociedade Brasileira de Diabetes · 2026
Diagnosis of diabetes mellitus — Brazilian guideline
- 04Open source ↗
U.S. Preventive Services Task Force · 2021
Screening for Prediabetes and Type 2 Diabetes
- 05Open source ↗
NIDDK · Current resource
Insulin Resistance and Prediabetes
- 06Open source ↗
Muniyappa et al. · 2008
Current approaches for assessing insulin sensitivity and resistance in vivo
- 07Open source ↗
Journal of Clinical Endocrinology & Metabolism · 2026
Biomarkers of insulin resistance and incident dysglycemia
Editorial record