Blood Glucose and Diabetes Screening: Tests, Thresholds, and Timing

Diabetes screening measures blood sugar in people who feel well, because type 2 diabetes can build silently for years before any symptom appears, and the numbers that matter are surprisingly clear once you know what each test is doing. ¹

Most people meet these tests not because anything is wrong, but because they ticked a box on a checkup form. The result comes back with a percentage or a millimole figure and very little explanation. This page is the plain version: what the three common tests measure, the thresholds that separate normal from prediabetes from diabetes, and what a borderline result should and should not make you do.

What diabetes screening is checking for

Screening looks for raised blood glucose before it announces itself. Type 2 diabetes can be present for years without symptoms, which is exactly why it is worth testing at-risk adults rather than waiting for thirst, fatigue, or blurred vision to arrive. ¹ By the time symptoms are obvious, blood sugar has often been high for a long time.

A screen is not a diagnostic test. It is testing someone without symptoms to catch a condition or its risk factors earlier than it would otherwise show up. An abnormal screen does not mean you have diabetes; it means the result is worth confirming. That distinction matters because the emotional weight of a single number is often heavier than the number deserves. For how this fits into a general appointment, see what a routine checkup actually includes.

The three common tests

Three blood tests do most of the work, and they measure different slices of the same picture.

• Fasting plasma glucose (FPG): a single snapshot taken after roughly 8 to 12 hours without food. It tells you where your blood sugar sits at rest. ²
• HbA1c (glycated haemoglobin): an average of your blood sugar over the past 2 to 3 months, read as a percentage or in mmol/mol. It needs no fasting, which makes it convenient, though certain conditions affecting red blood cells can distort it. ³
• Oral glucose tolerance test (OGTT): you drink a measured glucose load and your blood is tested before and two hours after. It shows how your body handles a sugar challenge and is often used in pregnancy or when other results are ambiguous. ²

No single test is universally best. Clinicians pick based on your situation, convenience, and sometimes order more than one when results disagree.

The numbers that define each band

Here is where the abstraction becomes concrete. Widely used international thresholds put diabetes at an HbA1c of 6.5% (48 mmol/mol) or above, or a fasting glucose of 7.0 mmol/L (126 mg/dL) or above. ³ Prediabetes, also called intermediate hyperglycaemia, sits in the band just below: an HbA1c of 5.7 to 6.4% (39 to 47 mmol/mol), or a fasting glucose in the impaired range. ² Below that is the normal range.

A few honest caveats. Exact cut-offs and the language around them vary between guideline bodies and countries, so the figures above are the common reference points rather than a single global rule. Units differ too: many countries report glucose in mmol/L, others in mg/dL, and HbA1c appears as both a percentage and as mmol/mol. If your result uses units you do not recognise, that is a reason to ask, not to panic.

What a borderline or abnormal result means

A single abnormal screen is not a diagnosis. Unless you have clear symptoms alongside a very high reading, one out-of-range result is typically repeated or confirmed with a second test before any label is applied. ² Recent illness, stress, a large meal close to a non-fasting sample, or even a lab variation can move a single number.

Prediabetes deserves its own paragraph because it is so often misread. It is a signal, not a sentence. It marks the stage where the body is struggling with glucose but has not crossed into diabetes, and it is the stage where changes to weight, diet, movement, and sleep have the most measurable leverage. Screening at this point is genuinely worthwhile partly because of what can still be changed. The harms of screening are real too: false positives, anxiety, and cascade testing where one borderline result triggers a chain of further tests. A good clinician weighs these against the benefit rather than testing reflexively. ⁴

How often, and who benefits most

There is no single correct interval. Many guideline bodies suggest screening at-risk adults periodically, often around every 3 years when results are normal, and more frequently for people with prediabetes or other risk factors such as a higher body weight, family history, or raised blood pressure. ² Frequent retesting of low-risk people adds little and can add harm.

Blood glucose rarely travels alone. It tends to cluster with blood pressure and cholesterol as part of overall cardiometabolic risk, which is why these are so often checked together. Reading them as a set, rather than one isolated number, gives a truer picture. For the lipid side of that cluster, see what a cholesterol panel measures and when to start.

The most useful frame is this: a glucose screen is one data point in a longer relationship with your own health, best read calmly, in context, and with someone who can tell you whether the next step is a repeat test, a lifestyle conversation, or simply nothing for now.

---

General information, not medical advice. Thresholds and intervals vary by country and guideline body; discuss your own results and risk with a qualified clinician.