Prediabetes: What the Diagnosis Means, and What Changes It

Quick answer

An HbA1c of 42 to 47 mmol/mol or a fasting plasma glucose of 5.5 to 6.9 mmol/l identifies raised risk of type 2 diabetes rather than a disease, and progression is not inevitable — in pooled cohort data, reversion to normal glucose levels was more common than progression. Ask your GP practice about referral to the NHS Diabetes Prevention Programme, and keep to annual HbA1c or fasting glucose testing rather than acting on one result alone.

A result in the prediabetes range is usually delivered as a warning, and it is often heard as a sentence. Neither framing is quite right. Non-diabetic hyperglycaemia, the term UK guidance uses, describes a glucose level that is above normal but below the threshold for diabetes. It identifies a group at raised risk. It does not say what will happen to any individual in that group.

The distinction matters because the evidence on what happens next is more encouraging than the word "prediabetes" suggests. Progression is common enough to take seriously and far from inevitable, and movement back into the normal range is well documented.

The UK thresholds

NICE defines high risk of type 2 diabetes, and the NHS defines non-diabetic hyperglycaemia, by either of two measurements:

HbA1c of 42 to 47 mmol/mol, which corresponds to 6.0 to 6.4 per cent on the older scale
Fasting plasma glucose of 5.5 to 6.9 mmol/l

Diabetes itself is diagnosed at an HbA1c of 48 mmol/mol or above, or a fasting plasma glucose of 7.0 mmol/l or above. Below 42 mmol/mol, or a fasting glucose below 5.5 mmol/l, NICE describes risk as moderate rather than high.

These are UK thresholds and they are not identical to those used elsewhere, which is one reason results interpreted against an American reference range can cause unnecessary alarm. NICE also recommends that people with non-diabetic hyperglycaemia have an HbA1c or fasting plasma glucose test at least once a year, so that progression is detected early if it happens.

A risk state, not a disease

Non-diabetic hyperglycaemia is not a disease with its own symptoms and natural history. It is a marker that glucose regulation is under strain. The clinically useful question is not "do I have it" but "which direction am I moving in, and what influences that".

It is also worth saying plainly that the raised risk is not only of diabetes. Glucose in this range tends to travel with central adiposity, raised blood pressure, an unfavourable lipid pattern and fatty liver, and those carry cardiovascular risk of their own. Addressing the cluster is more useful than watching one number.

What the progression data actually show

A pooled analysis of nineteen prospective cohort studies, published in The Lancet Global Health in 2025, followed 76,092 people for a median of almost ten years. Among those with prediabetes at the start, 12.5 per cent developed type 2 diabetes within ten years, while 36.1 per cent returned to normoglycaemia.

Those figures deserve care rather than celebration. They are averages across varied populations and definitions, and the picture changed considerably with baseline glucose. In the highest quartile of fasting plasma glucose, progression rose to 16.1 per cent and reversion fell to 13.4 per cent. Older age, male sex, overweight and obesity, a higher waist-to-height ratio and low HDL cholesterol all shifted the odds unfavourably. Where you sit within the range, and what else is going on metabolically, matters more than the label.

What the evidence shows reduces risk

The strongest evidence concerns weight and activity. In the US Diabetes Prevention Program, 3,234 adults with raised fasting and post-load glucose were randomised to placebo, metformin or an intensive lifestyle programme aiming for 7 per cent weight loss and at least 150 minutes of activity a week. Over an average of 2.8 years, diabetes incidence was 11.0 cases per 100 person-years on placebo, 7.8 with metformin and 4.8 with the lifestyle programme: reductions of 31 per cent and 58 per cent respectively. In long-term follow-up the advantage narrowed but persisted, with diabetes delayed by 27 per cent in the lifestyle group at fifteen years.

A 2024 meta-analysis in the American Journal of Clinical Nutrition pooled 44 randomised trials of lifestyle weight-loss interventions in people with prediabetes, covering 14,742 participants. Progression to type 2 diabetes fell by around 41 per cent, and regression to normal glucose levels rose by around half. The dose-response analysis is the part most worth knowing: across the 1 to 9 per cent range studied, each additional 1 per cent of body weight lost was associated with further reduction in progression and further increase in reversion, in a broadly linear relationship. The authors rated the certainty moderate and noted signs of publication bias, so the true effect may be smaller than the pooled figures suggest.

The practical reading is that modest, sustained change counts, and that there is no single threshold of weight loss below which effort is wasted.

The NHS Diabetes Prevention Programme

England has a national programme for exactly this group. The Healthier You NHS Diabetes Prevention Programme is a nine-month behaviour-change programme delivered either in groups or digitally, covering weight, diet and physical activity. People are identified through the NHS Health Check, through GP testing or through the Diabetes UK risk tool, and referred by their practice. NHS England states that completing the programme cuts the risk of developing type 2 diabetes by more than a third.

It is free, it is evidence-based, and it is underused. If you have a result in this range, asking your GP practice about referral is a reasonable first step regardless of what else you do.

What this means for you

A result of 44 mmol/mol is information, not a verdict. It warrants a conversation about weight, activity, sleep, blood pressure, lipids and liver, an annual repeat test, and an honest look at where in the range you sit. It does not warrant panic, and it does not warrant a private supplement protocol. If your result is at the upper end, or if it has risen over successive tests, that changes the urgency rather than the approach.