Insulin Resistance: Reading the Markers Before a Diabetes Diagnosis

Insulin resistance can exist with a normal HbA1c. But fasting insulin, HOMA-IR and a lipid ratio cannot reliably diagnose it on their own. The useful question is whether additional testing would change your care. 1 2

What HbA1c tells you, and what it does not

Insulin resistance means tissues respond less effectively to insulin. The pancreas may compensate by producing more, keeping glucose in range for a time. Progression to prediabetes or diabetes is not inevitable. 1

HbA1c reflects average glucose exposure over roughly three months, not insulin sensitivity. There is no fixed ten-to-fifteen-year countdown, and a normal result is not evidence that hidden damage is already established. 1

HbA1c remains the preferred NZ screening test for most people without symptoms. Significant iron deficiency, altered red-cell turnover and some haemoglobin conditions can make it unreliable. 3

When HbA1c is unsuitable, fasting glucose or an oral glucose tolerance test may be more appropriate. These assess glucose regulation, rather than directly measuring insulin resistance. 3

The three markers: useful clues, important limits

Fasting insulin measures the concentration of insulin circulating after fasting. It is not a direct measurement of pancreatic workload or a stand-alone diagnostic test. 2

There is no universally accepted optimal range of 3 to 6 mIU/L. Insulin assays are not standardised, so interpretation needs the laboratory method, glucose result and clinical context. 2

A result of 15 or 18 mIU/L should not automatically be labelled dangerous. Equally, a result inside a laboratory reference interval does not prove normal insulin sensitivity. 2

HOMA-IR estimates insulin resistance from fasting insulin and glucose. The commonly quoted calculation is the HOMA1 approximation, not the more complex HOMA2 model. 4

HOMA1-IR = fasting insulin (mIU/L) × fasting glucose (mmol/L) ÷ 22.5. 4 5

Use insulin and glucose sampled together after fasting. Do not insert an insulin value reported in pmol/L into this formula. Confirm units and sample requirements with the laboratory or clinician. 4

There is no universal HOMA-IR boundary separating excellent, mild and significant resistance. Assay differences and model assumptions matter; online cut-offs are not a substitute for assessment. 4

Triglyceride-to-HDL ratio uses two lipid-panel results: triglycerides divided by HDL cholesterol. If both are in mmol/L, triglycerides of 1.8 and HDL of 1.2 give a ratio of 1.5. 5

That number is not a diagnosis. Associations with insulin resistance differ between populations and sexes. In one UK study, the association was not significant in South Asian women. 5

Do not import thresholds from US results reported in mg/dL. Triglycerides and cholesterol use different conversion factors, so the numerical ratio changes between unit systems. 5

What the markers actually tell you together

These are not three independent confirmations. HOMA-IR already contains fasting insulin, while the lipid ratio is a separate but imperfect clue. Agreement does not remove their limitations. 4 5

Consider a hypothetical fasting glucose of 5.2 mmol/L, HbA1c of 36 mmol/mol and fasting insulin of 18 mIU/L. The HOMA1 calculation is approximately 4.2. 4

The glucose results are below NZ prediabetes thresholds. The insulin-derived estimate may prompt discussion, but it does not prove the pancreas is working five times harder than it should. 2 3

Interpret the full picture: glucose results, blood pressure, lipids, relevant medicines and family history. A favourable ratio should not override other cardiovascular or diabetes risk factors. 2 3

What to discuss with your GP in New Zealand

NZ screening is risk-based, not simply a fasting-insulin request for everyone over 35. Relevant factors include previous gestational diabetes, prediabetes, family history, ethnicity and clinical features such as PCOS. 3

People with prediabetes or previous gestational diabetes generally need annual HbA1c testing. Other screening intervals depend on risk. Ask when your next assessment is due. 3

Routine insulin testing is generally not recommended for most people at risk of diabetes. A GP declining it may be following evidence-based guidance, rather than overlooking an early-warning test. 2

Useful questions include:

Is my HbA1c reliable in my circumstances?
Would fasting glucose or a glucose tolerance test add useful information?
Would fasting insulin change the management plan?
What should we monitor, and when should we review it?

What moves metabolic health in the right direction

Combine strength and aerobic activity. Both can improve insulin sensitivity. Build towards strengthening on at least two days and 150 minutes of moderate activity weekly. Start below that if needed. 6

Bodyweight squats, wall push-ups, resistance bands and weights are options. Progress gradually rather than treating a particular training frequency as a guaranteed insulin-lowering dose. 6

Try walking after meals. In 41 adults with type 2 diabetes, ten-minute walks after each main meal improved post-meal glucose compared with one daily 30-minute walk. Each intervention lasted two weeks. 7

This supports a practical habit, not a promise to reverse insulin resistance. It does not establish the same effect from one walk, lower fasting insulin or lasting diabetes prevention. 7

Build balanced meals. Include protein foods such as beans, tofu, fish, eggs or yoghurt. Choose fibre-rich carbohydrates such as oats, legumes, fruit, kūmara and wholegrain bread. 6

Food quality and portions both matter. There is no need to label all bread harmful or prescribe a lean-mass protein target as treatment for insulin resistance. 6

Include sleep in the plan. Protect sufficient sleep alongside food and activity changes, rather than expecting a particular number of hours to normalise a blood result within a week. 6

Track progress without overreading a scan

Where weight loss is appropriate, modest sustained loss can help reduce diabetes risk. Prevention programmes have used goals around 5% to 7% of starting weight, alongside food and activity changes. 1

BIA estimates body composition. Its visceral-fat output is not a direct measurement. A validation study found limited agreement with CT, although its findings apply to the single-frequency device studied. 8

Do not use a changing BIA score to confirm insulin resistance or prove that fat loss came specifically from around the organs. Device estimates cannot replace clinical assessment. 8

What to do this week

Gather your latest HbA1c, glucose and lipid results, including dates and units.
Note relevant family history, medicines and previous gestational diabetes for your GP discussion.
Choose one manageable activity change, such as a short post-meal walk.
Review three days of meals for protein foods, fibre-rich choices and portions.
Agree on follow-up rather than repeatedly ordering insulin tests without a clinical question.

Seek prompt medical assessment for new excessive thirst, frequent urination or unexplained weight loss. These need glucose-based evaluation, not a wait for a fasting-insulin appointment. 9

For help turning the lifestyle steps into a workable routine, nutrition coaching can support implementation. Diagnostic assessment and treatment decisions remain with your healthcare team.

References

1.National Institute of Diabetes and Digestive and Kidney Diseases. Insulin Resistance & Prediabetes.[Patient guidance on insulin resistance and prevention.] Source

2.Sacks DB, et al. Guidelines and Recommendations for Laboratory Analysis in the Diagnosis and Management of Diabetes Mellitus. Diabetes Care. 2023;46(10):e151-e199.[Professional guidance on appropriate testing and laboratory limitations.] doi:10.2337/dci23-0036

3.New Zealand Society for the Study of Diabetes. Screening for diabetes in asymptomatic adults.[NZ screening and HbA1c interpretation guidance.] Source

4.University of Oxford, Radcliffe Department of Medicine. Frequently Asked Questions: HOMA2 Calculator.[Model-developer explanation of calculations, sample requirements and interpretation.] Source

5.Mostafa SA, et al. The Association of the Triglyceride-to-HDL Cholesterol Ratio with Insulin Resistance in White European and South Asian Men and Women. PLoS ONE. 2012;7(12):e50931.[Cross-sectional study demonstrating population and sex limitations.] doi:10.1371/journal.pone.0050931

6.Ministry of Health. Eating and Activity Guidelines for New Zealand Adults. Updated 2020. Wellington: Ministry of Health.[General nutrition and physical activity guidance.] Source

7.Reynolds AN, Mann JI, Williams S, Venn BJ. Advice to walk after meals is more effective for lowering postprandial glycaemia in type 2 diabetes mellitus than advice that does not specify timing: a randomised crossover study. Diabetologia. 2016;59(12):2572-2578.[Short crossover trial in 41 adults with type 2 diabetes.] doi:10.1007/s00125-016-4085-2

8.Measurement of visceral fat and abdominal obesity by single-frequency bioelectrical impedance and CT: a cross-sectional study. BMJ Open. 2021;11:e048221.[Device-specific validation, not a universal estimate of BIA accuracy.] doi:10.1136/bmjopen-2020-048221

9.New Zealand Society for the Study of Diabetes. Screening for diabetes in symptomatic adults.[Symptoms requiring clinical assessment.] Source

Quick answers

Can I have insulin resistance with a normal HbA1c?

Yes. Your pancreas may compensate by producing more insulin while glucose stays in range. HbA1c measures glucose exposure, not insulin sensitivity, and there is no fixed timeline to diabetes.

What is a good fasting insulin level?

There is no universally accepted optimal range. Interpretation depends on the laboratory method, glucose result and clinical context. A single reading cannot reliably diagnose or exclude insulin resistance.

How do I calculate HOMA-IR?

For HOMA1-IR, multiply fasting insulin in mIU/L by fasting glucose in mmol/L, then divide by 22.5. Use samples taken together. This is an estimate, not HOMA2, and universal diagnostic cut-offs do not apply.

How do I calculate my triglyceride-to-HDL ratio?

Divide triglycerides by HDL cholesterol, using mmol/L for both. Do not apply cut-offs based on mg/dL. The ratio is a contextual clue, not a diagnosis or a substitute for reviewing the full lipid panel.

Which lifestyle change lowers fasting insulin fastest?

There is no reliable fastest option for everyone. Start with regular activity, balanced meals and sufficient sleep. Short-term post-meal glucose improvements do not guarantee lower fasting insulin.