Insulin resistance: the 13-year head start standard testing misses.

Insulin resistance is a state in which your muscle, liver and fat cells respond poorly to insulin, forcing the pancreas to produce more of it to keep blood glucose normal. The critical feature, and the reason it goes undetected for so long, is that it works. For years, sometimes more than a decade, the extra insulin succeeds at holding your blood sugar in the normal range. Your fasting glucose reads fine. Your HbA1c reads fine. The only thing that is abnormal is the amount of insulin required to produce that normal result, and no standard Australian health check measures it.

The key facts, up front

• Insulin sensitivity starts declining up to 13 years before diagnosis. In the Whitehall II cohort (Lancet, 2009), insulin sensitivity was already falling 13 years before type 2 diabetes was diagnosed, while fasting glucose only began its sharp rise in the final 3 years.
• A normal glucose does not mean normal metabolism. Glucose is the last domino to fall, not the first. By the time HbA1c crosses into the prediabetes range, beta-cell function has typically already been declining for years.
• Roughly 1 in 7 Australian adults with diabetes doesn't know it. AIHW data for 2022–24 estimate 6.5% of Australian adults (about 1.3 million people) have diabetes, of whom around 0.9 percentage points were newly detected on testing rather than previously diagnosed.
• It is one of the most reversible risk states in medicine. Structured lifestyle intervention cut progression to type 2 diabetes by 58% in both the US Diabetes Prevention Program and the Finnish Diabetes Prevention Study.
• The tests are cheap and the maths is simple. Fasting insulin plus fasting glucose from a single blood draw gives you HOMA-IR. Triglycerides and HDL from the same panel give you the TG/HDL ratio.

What actually gets measured: and what it means

Insulin resistance has no single perfect test outside a research laboratory. The reference standard is the hyperinsulinaemic-euglycaemic clamp, described by DeFronzo and colleagues in 1979: a multi-hour procedure involving continuous infusions and repeated sampling. It is precise, and it is entirely impractical for routine care. What clinical practice uses instead is a small set of accessible surrogates, each with a defined role in the HodieLabs Evidence-Based Clinical Library.

• Fasting insulin, the earliest single signal. It reflects how hard the pancreas is working at rest. Our interpretation bands: 2.0–5.0 mIU/L indicates strong insulin sensitivity, 5.1–8.0 mIU/L is within a healthy range, 8.1–12.0 mIU/L suggests early insulin resistance, and above 12.0 mIU/L is high.
• HOMA-IR. The calculated index published by Matthews and colleagues in Diabetologia in 1985. In Australian units: fasting insulin (mIU/L) × fasting glucose (mmol/L) ÷ 22.5. Below 1.5 reflects optimal sensitivity; 1.5–2.5 is borderline; above 2.5 is widely treated across the literature as consistent with insulin resistance.
• TG/HDL ratio. The most underused marker on a standard lipid panel, requiring no extra blood draw. Below 1.0 (mmol/L units) is favourable; above roughly 1.5 is associated with the small, dense LDL pattern and predicts metabolic syndrome even when LDL-cholesterol looks acceptable.
• HbA1c, a genuinely excellent test, but a late one. It reflects average glucose over 2–3 months, which by definition means it only moves once glucose control has already been lost. ADA thresholds: below 5.7% normal, 5.7–6.4% prediabetes, 6.5% or above diabetes.

What the clinical evidence actually shows

Three things need separate evidence: that insulin resistance precedes disease, that it predicts hard outcomes, and that intervening changes those outcomes. All three have been tested.

• Whitehall II (Tabák et al. Lancet 2009): by working backwards from the point of diagnosis in a cohort of British civil servants followed for over a decade, the authors mapped the natural history precisely. Insulin sensitivity was declining from about 13 years before diagnosis; fasting glucose rose gradually and then steeply only in the final ~3 years; beta-cell function rose in compensation before collapsing. This is the single clearest demonstration that the metabolic problem long precedes the glucose problem.
• AusDiab (Barr et al. Circulation 2007): in over 10,000 Australian adults, impaired fasting glucose and impaired glucose tolerance, both pre-diabetic states, were associated with increased cardiovascular and all-cause mortality. Metabolic risk is not binary and does not begin at the diabetes threshold.
• Diabetes Prevention Program (Knowler et al. NEJM 2002): 3,234 adults with impaired glucose tolerance randomised to intensive lifestyle intervention, metformin or placebo. Lifestyle reduced the incidence of type 2 diabetes by 58%; metformin by 31%. Lifestyle beat the drug, decisively.
• Finnish Diabetes Prevention Study (Tuomilehto et al. NEJM 2001): an independent trial in a different population reached an almost identical figure, a 58% reduction, which is why this result is treated as robust rather than a single-trial fluke.
• Da Qing 30-year follow-up (Gong et al. Lancet Diabetes & Endocrinology 2019): the longest follow-up of any diabetes prevention trial. Three decades after a six-year lifestyle intervention in Chinese adults with impaired glucose tolerance, the intervention group had fewer cardiovascular events, lower cardiovascular and all-cause mortality, and an average increase in life expectancy of roughly 1.4 years. The benefit of acting early persisted long after the intervention ended.
• DiRECT (Lean et al. Lancet 2018): even established type 2 diabetes is not necessarily permanent. In a primary-care weight-management programme, 46% of intervention participants were in remission at 12 months, versus 4% of controls; remission tracked closely with the amount of weight lost.

Where the evidence is weaker: and we'll say so

Being evidence-based means being clear about what the evidence does not support. Three honest caveats apply to insulin testing:

• Insulin assays are not fully standardised. Different laboratories using different immunoassays can return meaningfully different values for the same sample. This is a real, documented limitation. The practical consequence: track your trend within one laboratory rather than comparing absolute numbers between providers.
• There is no society-endorsed diagnostic cut-point for fasting insulin. The ADA does not define diabetes or prediabetes by insulin level, and neither do Australian guidelines. The "optimal" bands used above are drawn from observational metabolic-health literature and are legitimate coaching targets, but they are not diagnostic thresholds, and we label them as such throughout the library.
• HOMA-IR is a surrogate, not the clamp. It correlates well at population level and reasonably at individual level, but it is a model, not a measurement. A single elevated HOMA-IR in an otherwise healthy person warrants a repeat test and clinical context, not alarm.

None of this makes the markers useless. It makes them what they are: early, cheap, directionally reliable signals that belong in a full metabolic picture alongside glucose, HbA1c, lipids and body composition, not isolated numbers to be over-interpreted.

The Australian gap

A standard Australian health check measures fasting glucose and HbA1c. Both are good tests. Both, by design, detect the problem at the stage where glucose control has already begun to fail. The last few years of a process that started more than a decade earlier. Fasting insulin is available through Australian pathology labs, but it is rarely ordered outside specific clinical indications and is often not Medicare-rebated without one. The result is a screening system well designed to diagnose diabetes and poorly designed to prevent it, in a country where AIHW data indicate roughly 1.3 million adults are living with diabetes and a meaningful share of those cases are only found when someone finally gets tested.

How testing works at HodieLabs

Fasting insulin, fasting glucose, HOMA-IR, HbA1c and the TG/HDL ratio are all measured or derived in every Health Blueprint at our Melbourne Preventative Health & Longevity Clinic. They are interpreted together, a fasting insulin of 11 mIU/L alongside a perfectly normal glucose tells a story that neither number tells alone, and read by a doctor in the context of your complete biomarker picture, including body composition and cardiovascular markers such as ApoB. Where results indicate early insulin resistance, the plan follows what the trials actually tested: resistance and aerobic training, reduced refined carbohydrate load, visceral fat reduction and sleep. Then a recheck at three months to confirm the trend is moving.

The bottom line

Type 2 diabetes is not an event; it is the visible end of a process that has been running for over a decade. Glucose and HbA1c tell you where that process has arrived. Fasting insulin and HOMA-IR tell you where it is heading. Imperfectly, but years earlier, and at a point where the evidence says intervention still works remarkably well. If your last check was "normal blood sugar" and nothing else, you have measured the outcome without measuring the cause.

References

1. Tabák AG, Jokela M, Akbaraly TN, et al. Trajectories of glycaemia, insulin sensitivity, and insulin secretion before diagnosis of type 2 diabetes: an analysis from the Whitehall II study. Lancet. 2009;373(9682):2215–2221.
2. Matthews DR, Hosker JP, Rudenski AS, et al. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985;28(7):412–419.
3. DeFronzo RA, Tobin JD, Andres R. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol. 1979;237(3):E214–E223.
4. Knowler WC, Barrett-Connor E, Fowler SE, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med. 2002;346(6):393–403.
5. Tuomilehto J, Lindström J, Eriksson JG, et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med. 2001;344(18):1343–1350.
6. Gong Q, Zhang P, Wang J, et al. Morbidity and mortality after lifestyle intervention for people with impaired glucose tolerance: 30-year results of the Da Qing Diabetes Prevention Outcome Study. Lancet Diabetes Endocrinol. 2019;7(6):452–461.
7. Lean MEJ, Leslie WS, Barnes AC, et al. Primary care-led weight management for remission of type 2 diabetes (DiRECT): an open-label, cluster-randomised trial. Lancet. 2018;391(10120):541–551.
8. Barr ELM, Zimmet PZ, Welborn TA, et al. Risk of cardiovascular and all-cause mortality in individuals with diabetes mellitus, impaired fasting glucose, and impaired glucose tolerance: the Australian Diabetes, Obesity, and Lifestyle Study (AusDiab). Circulation. 2007;116(2):151–157.
9. Reaven GM. Banting Lecture 1988. Role of insulin resistance in human disease. Diabetes. 1988;37(12):1595–1607.
10. American Diabetes Association. Standards of Care in Diabetes, 2024. Diabetes Care. 2024;47(Suppl 1).
11. Australian Institute of Health and Welfare. Diabetes: Australian facts. AIHW, Canberra (2022–24 prevalence data).

This article is general information, not medical advice. Speak with your GP about what testing and care is appropriate for you.