OGTT vs HbA1c: Which Test Finds Dysglycaemia First?

Disclaimer: This article is for educational and research purposes only. It does not constitute medical advice. Consult a qualified healthcare professional before making any health-related decisions based on blood test results.

Three tests sit at the centre of glucose metabolism assessment in Australia: fasting plasma glucose, HbA1c, and the oral glucose tolerance test (OGTT). All three are used to diagnose diabetes and prediabetes. All three are ordered in Australian general practice. And all three measure fundamentally different things, a distinction that matters enormously when metabolic dysfunction is early, subtle, or of an unusual type.

The failure mode is assuming the tests are interchangeable. They are not. Each captures a different dimension of glucose regulation, and each has blind spots the others do not share. Understanding those blind spots is what separates a metabolic panel that catches problems early from one that returns a clean result while dysfunction progresses undetected.

What Each Test Actually Measures

Fasting Plasma Glucose

Fasting glucose measures the concentration of glucose in plasma after a minimum eight-hour fast. It reflects the balance between hepatic glucose output, the liver's rate of releasing stored glucose into circulation, and baseline insulin-mediated suppression of that output.

Because it is a snapshot taken under controlled, fasted conditions, it is highly sensitive to hepatic insulin resistance. A liver that is resistant to insulin continues producing glucose at elevated rates even when blood glucose is already adequate. This manifests as elevated fasting glucose before postprandial glucose has meaningfully deteriorated.

The diagnostic thresholds in Australia follow WHO and RACGP guidance: normal is below 6.0 mmol/L, impaired fasting glucose (IFG) sits between 6.0 and 6.9 mmol/L, and diabetes is diagnosed at 7.0 mmol/L or above on two separate occasions.

HbA1c

HbA1c measures the proportion of haemoglobin that has been glycated, chemically modified by glucose, over the preceding 90 to 120 days, which corresponds to the lifespan of a red blood cell. It is not a snapshot; it is a weighted average of blood glucose exposure over that entire window, with more recent glucose levels weighted more heavily.

The Australian reference ranges classify normal as below 41 mmol/mol (below 5.9%), prediabetes as 41–49 mmol/mol, and diabetes as 50 mmol/mol or above. For a detailed discussion of functional targets below the diagnostic threshold, see HbA1c optimal ranges and functional targets.

HbA1c has significant practical advantages: it requires no fasting, can be taken at any time of day, is less subject to acute illness-related fluctuation than single glucose readings, and provides a reliable longitudinal signal for monitoring interventions. These properties made it attractive as a screening tool and drove its incorporation into diagnostic criteria by the International Expert Committee in 2009.

Oral Glucose Tolerance Test

The OGTT is a dynamic, provocative test. It does not assess glucose under resting or fasted conditions alone, it assesses the body's capacity to handle an acute glucose load. The standard 75-gram protocol involves an eight-hour overnight fast, a baseline fasting glucose measurement, consumption of a 75-gram glucose solution over five minutes, and a two-hour post-load glucose measurement. In the gestational context, a three-hour, 100-gram protocol is also used.

The two-hour value reveals something neither fasting glucose nor HbA1c can show: the efficiency of the entire glucose disposal system under challenge conditions. This encompasses insulin secretion from beta cells in response to the acute stimulus, peripheral glucose uptake in skeletal muscle, hepatic glucose suppression, and incretin hormone activity. The OGTT is the only standard clinical test that directly measures postprandial glucose handling.

Where HbA1c Misses Early Dysglycaemia

HbA1c's most significant limitation is structural: it measures the average glucose signal over time, which means intermittent or postprandial elevations can be absorbed into a result that still appears normal.

Consider the mechanism. A person with early impaired glucose tolerance may have normal fasting glucose and a normal HbA1c. Their fasting glucose is low enough, and their overnight glucose levels are stable enough, that the 90-day average does not reflect the abnormality. What is abnormal is the two-hour response to a glucose load: their glucose climbs above 7.8 mmol/L after a meal and remains elevated for longer than it should before returning to baseline. That pattern, elevated postprandial glucose, can persist for years while HbA1c remains under 41 mmol/mol.

This is not a theoretical concern. Studies comparing OGTT and HbA1c in matched populations consistently show that the two tests identify different individuals as dysglycaemic, with modest overlap. The OGTT classifies a meaningfully larger proportion of at-risk individuals, particularly those with isolated postprandial glucose dysregulation, impaired glucose tolerance (IGT), that does not affect the fasting or average glucose signal.

HbA1c also has biological confounders that can generate false reassurance. In individuals with haemolytic anaemia, thalassaemia trait, iron-deficiency anaemia, or haemoglobin variants, red cell turnover is altered and HbA1c does not accurately reflect average glucose. In these populations, a normal HbA1c is not reliable evidence of normal glucose metabolism. The OGTT is not subject to these haematological confounders.

For the relationship between insulin resistance and the glucose markers that detect it earliest, fasting insulin, the most underordered metabolic test in Australia covers how fasting insulin can be substantially elevated while both HbA1c and fasting glucose remain in the normal range.

The 1-Hour Glucose Threshold: An Emerging Standard

The standard OGTT reads glucose at two hours. A growing body of evidence suggests the one-hour post-load reading may be more informative, and may become clinically standard.

Research published in Diabetes/Metabolism Research and Reviews reviewed data across multiple large cohort studies and found that a one-hour post-load plasma glucose of 8.6 mmol/L (155 mg/dL) or above, even in individuals whose two-hour value was below the 7.8 mmol/L threshold for impaired glucose tolerance, predicted progression to type 2 diabetes, microvascular disease, and mortality. The one-hour value captured individuals at genuine metabolic risk who would be classified as normal under current two-hour-only criteria. Read the study at PubMed.

The biological rationale is plausible. The one-hour reading reflects the peak of first-phase insulin secretion, the immediate beta cell response to rising glucose. Early beta cell dysfunction manifests as a blunted or delayed first-phase response, allowing glucose to overshoot at one hour before second-phase insulin eventually brings it down. By two hours, glucose may have returned to an apparently normal range, leaving no trace of the dysfunction in the standard result.

The International Diabetes Federation released a position statement in 2024 recommending the one-hour plasma glucose threshold of 8.6 mmol/L as a criterion for diagnosing intermediate hyperglycaemia, a formal step toward incorporating the one-hour reading into standard diagnostic practice. This threshold is not yet embedded in RACGP or NHMRC guidelines in Australia, but practitioners ordering OGTTs are increasingly noting and acting on the one-hour value when present.

A 2025 study published in Scientific Reports with over 21 years of follow-up in 2,347 participants found that postprandial 2-hour glucose tolerance was independently associated with diabetes diagnosis, diabetes mortality, and cardiovascular mortality, reinforcing the clinical weight of dynamic glucose response measurements beyond fasting and HbA1c averages. View the PMC record.

Standard OGTT Protocol in Australia

The 75-gram OGTT in Australia is performed as a Medicare-funded test under specific clinical indications. The protocol is standardised:

• Pre-test requirements: Eight to fourteen hours of fasting. Normal carbohydrate intake (at least 150 grams per day) for three days prior. No smoking on the test morning. The patient should be ambulatory and not acutely unwell.
• Preparation: A fasting blood sample is drawn on arrival.
• Glucose load: A 75-gram glucose solution (typically 300 mL of commercially prepared glucose drink) is consumed within five minutes.
• Timing: Blood is drawn again at two hours. Some laboratories or clinicians also record the one-hour value.
• Interpretation: Two-hour plasma glucose below 7.8 mmol/L is normal; 7.8–11.0 mmol/L indicates impaired glucose tolerance; 11.1 mmol/L or above is diagnostic for diabetes (confirmed by a second test or in the presence of symptoms).

The test takes roughly two and a half hours including wait time. Patients should remain seated and avoid strenuous activity during the test window.

OGTT in the Gestational Context

Gestational diabetes mellitus (GDM) is one of the most common pregnancy complications in Australia, affecting approximately 15–17% of pregnancies under current diagnostic criteria. The OGTT is the definitive diagnostic test for GDM in Australia.

The Australasian Diabetes in Pregnancy Society (ADIPS) and NHMRC recommend universal screening with a 75-gram OGTT between 24 and 28 weeks of gestation in women not already known to have diabetes. Women with risk factors (prior GDM, pre-pregnancy BMI above 30, polycystic ovary syndrome, family history of type 2 diabetes, or from high-risk ethnic backgrounds) may be tested earlier, typically in the first trimester.

The diagnostic thresholds for GDM differ from the non-gestational OGTT. Under Australian criteria (based on the HAPO study and ADIPS 2014 guidelines), GDM is diagnosed if any one value meets or exceeds: fasting 5.1 mmol/L, one-hour 10.0 mmol/L, or two-hour 8.5 mmol/L. These are lower thresholds than the standard diagnostic criteria and reflect the particular sensitivity of fetal development to maternal glucose levels throughout the day, including postprandial elevations that would not meet the threshold for diagnosing GDM in a non-pregnant adult.

HbA1c is not used to diagnose GDM. Its 90-day averaging window and the haemodilution and increased red cell turnover of pregnancy make it an unreliable diagnostic tool in this context. HbA1c may be used in early pregnancy to screen for pre-existing, undiagnosed diabetes, a distinct clinical question from GDM diagnosis, but it does not replace the OGTT for the 24–28-week assessment.

Postpartum, ADIPS recommends a repeat 75-gram OGTT at six to twelve weeks after delivery for all women diagnosed with GDM, and annually thereafter given the substantially elevated lifetime risk of progression to type 2 diabetes.

When to Use Which Test

The tests are not competing options from which one should be selected. They are complementary tools, each suited to specific clinical questions.

Use HbA1c for:

• Longitudinal monitoring of glycaemic control in individuals already diagnosed with diabetes or prediabetes
• Population screening in individuals with standard metabolic risk factors where a non-fasting test is practical
• Assessing the effectiveness of dietary, lifestyle, or pharmacological interventions over a 3-month period
• Situations where the patient cannot tolerate a two-hour test

Use fasting glucose for:

• Initial metabolic triage in a standard panel
• Detecting hepatic insulin resistance, which manifests early as elevated fasting glucose
• Pairing with fasting insulin for HOMA-IR calculation, discussed in detail at HOMA-IR and insulin resistance

Use the OGTT when:

• HbA1c and fasting glucose are in the normal range but symptoms, risk factors, or clinical suspicion suggest early dysglycaemia
• Isolated impaired glucose tolerance is suspected (normal fasting glucose but possible postprandial dysfunction)
• Diagnosing or ruling out gestational diabetes (mandatory)
• Haematological conditions make HbA1c unreliable
• Assessing beta cell function and insulin secretory capacity under challenge
• A clinician wants to capture the full glucose disposal profile, including the emerging one-hour reading

Functional Interpretation: What to Look For Beyond the Reference Range

Thinking about these tests in terms of binary pass/fail thresholds misses a substantial portion of their information content.

For HbA1c, cardiovascular risk data shows a dose-response relationship beginning around 31–33 mmol/mol, well below the prediabetes threshold of 41 mmol/mol. A result of 38–40 mmol/mol is technically "normal" but warrants further metabolic investigation, particularly pairing with fasting insulin.

For fasting glucose, a reading between 5.0 and 5.5 mmol/L in the presence of other metabolic risk factors (elevated triglycerides, low HDL, central adiposity, elevated blood pressure) suggests early insulin resistance even though it sits comfortably within the reference range.

For the OGTT, the pattern of the response matters as much as the two-hour endpoint. A glucose value that peaks very high at one hour (above 8.6 mmol/L) and then returns to normal by two hours should not simply be dismissed because the final reading is technically below 7.8 mmol/L. That pattern indicates impaired first-phase insulin secretion, an early signal of beta cell strain, even in the absence of a diagnostic IGT result under current criteria.

The combination of normal fasting glucose, normal HbA1c, and an abnormal one-hour OGTT reading represents exactly the clinical scenario where current standard testing would return a clean result and early dysglycaemia would go undetected. That gap is the case for why dynamic testing, at least in higher-risk individuals, adds genuine diagnostic value that static and averaged tests cannot replicate.

Key Takeaways

• Fasting glucose measures hepatic glucose output under resting conditions; HbA1c measures 90-day average glucose exposure; the OGTT measures the dynamic glucose disposal response to an acute 75-gram load, they are not interchangeable
• HbA1c misses isolated postprandial glucose dysregulation (impaired glucose tolerance) because intermittent elevations can be absorbed into a still-normal average
• HbA1c is unreliable in haemolytic anaemia, thalassaemia, iron deficiency, and haemoglobin variants, the OGTT is the appropriate alternative
• The one-hour post-load glucose reading of 8.6 mmol/L (155 mg/dL) is an emerging threshold: individuals exceeding it even with a normal two-hour result show elevated risk of progression to type 2 diabetes and cardiovascular mortality
• GDM in Australia is diagnosed via OGTT only; HbA1c is not used for this purpose; diagnostic thresholds are lower than standard non-pregnant criteria (fasting ≥5.1, 1-hour ≥10.0, 2-hour ≥8.5 mmol/L)
• A normal HbA1c and fasting glucose do not exclude early dysglycaemia; OGTT adds diagnostic value in high-risk individuals, those with symptoms of glucose dysregulation, and wherever postprandial glucose handling is the clinical question