Prediabetes by the Numbers – Metabolic Media

Disclosure: Metabolic Media is an independent educational publisher. We may have a financial affiliation with some entities featured on this site. Rank positions are determined by our published methodology. We do not provide medical care, diagnoses, or treatment.

The short answer: The ADA’s 2026 Standards define prediabetes as an A1C of 5.7–6.4%, fasting glucose of 100–125 mg/dL, or a 2-hour OGTT glucose of 140–199 mg/dL. CDC (updated Sept 2026) estimates 115.2 million U.S. adults have it. In the Diabetes Prevention Program trial, diabetes incidence was 11.0 per 100 person-years on placebo and 4.8 with an intensive lifestyle program.

This explainer puts the main prediabetes numbers in one place, each with its source, its date, and the kind of study behind it. It summarizes research and guidelines; it doesn’t diagnose anyone or recommend treatment.

What numbers define prediabetes?

The table below shows the ranges in the American Diabetes Association’s Standards of Care in Diabetes—2026, Section 2 (criteria for nonpregnant adults).

Test	Prediabetes range	Diabetes range
A1C	5.7–6.4% (39–47 mmol/mol)	6.5% or higher (48 mmol/mol or higher)
Fasting plasma glucose	100–125 mg/dL (5.6–6.9 mmol/L), called impaired fasting glucose	126 mg/dL or higher (7.0 mmol/L or higher)
2-hour glucose, 75-g oral glucose tolerance test (OGTT)	140–199 mg/dL (7.8–11.0 mmol/L), called impaired glucose tolerance	200 mg/dL or higher (11.1 mmol/L or higher)

Source: ADA Standards of Care in Diabetes—2026, Tables 2.1 and 2.2. As of the 2026 Standards.

The ADA adds a caveat: for all three tests, risk is continuous. It extends below the lower limit of each range and becomes disproportionately greater toward the higher end. These cutoffs are conventions along a sliding scale, not an on/off switch.

The ADA also notes that the World Health Organization and several other groups set the lower fasting-glucose limit at 110 mg/dL, while the ADA adopted 100 mg/dL in 2003.

Why can one result look different from another?

A1C and glucose tests measure different things and vary for different reasons. The ADA’s Standards describe higher within-person variability for glucose tests, and note that A1C can be affected by anemia, hemoglobin variants, pregnancy, and other conditions that change red blood cell turnover.

For diagnosing diabetes, the Standards say that when two different tests disagree, the result above the diagnostic cut point is repeated, and that results close to a threshold prompt a repeat test in 3–6 months. That describes how the guideline works, not what any one reader should do.

Prediabetes usually has no symptoms, which is one reason it’s defined by lab values. For how cells respond to insulin behind those values, see Insulin Resistance, Explained.

How many people have prediabetes?

CDC’s National Diabetes Statistics Report (page updated Sept 2026) estimates that 115.2 million U.S. adults aged 18 and older have prediabetes. The same page estimates 31.3 million adults aged 65 and older (52.1% of that age group) have it.

For context, the page estimates 40.1 million people in the U.S. had diagnosed or undiagnosed diabetes (2023 estimate), including 11.0 million adults whose diabetes was undiagnosed.

These are CDC estimates, not a head count.

Who gets tested, and how often?

The ADA’s 2026 Standards recommend testing all adults starting at age 35. They recommend testing earlier for adults with overweight or obesity (BMI 25 or higher, or 23 or higher in people of Asian ancestry) who have one or more risk factors, such as a first-degree relative with diabetes, high blood pressure, cardiovascular disease, or physical inactivity.

For people whose results are normal, the Standards call for repeat testing at a minimum of every three years; for people with prediabetes, yearly testing. This is the guideline as written, as of the 2026 Standards, not advice for any one person.

What did the Diabetes Prevention Program find?

The Diabetes Prevention Program (DPP) was a randomized trial published in the New England Journal of Medicine in 2002 (PubMed). It was supported by NIDDK and other NIH institutes and U.S. agencies, the American Diabetes Association, Bristol-Myers Squibb, and Parke-Davis; Lipha Pharmaceuticals supplied the metformin and placebo. One author reported owning stock in Bristol-Myers Squibb, which sells metformin in the U.S. The figures here come from the published abstract.

Who took part: 3,234 adults without diabetes but with elevated fasting and post-load glucose. Mean age was 51, mean BMI was 34.0, 68% were women, and 45% were from minority groups.
What was compared: placebo, a medication arm (metformin), and an intensive lifestyle program with goals of at least 7% weight loss and at least 150 minutes of physical activity per week.
How long: an average of 2.8 years of follow-up.

In the trial, diabetes developed at 11.0 cases per 100 person-years with placebo, 7.8 with metformin, and 4.8 with the lifestyle program. Expressed as relative reductions versus placebo, the authors report 58% for lifestyle and 31% for metformin. Over a three-year period, the authors estimate that 6.9 people would need to participate in the lifestyle program, or 13.9 receive metformin, for one fewer case of diabetes to occur.

These are results for a selected, high-risk group. Whether any approach fits a particular person is a decision for that person and their clinician.

What did the long-term follow-up show?

NIDDK also funded the Diabetes Prevention Program Outcomes Study (DPPOS), which followed participants after the original trial. In these follow-ups, all participants were offered lifestyle training, so they are not new randomized comparisons between the original groups.

15 years (Lancet Diabetes & Endocrinology, 2015): 2,776 participants were followed. Cumulative diabetes incidence at year 15 was 62% in the placebo group, 56% in the metformin group, and 55% in the lifestyle group. The authors report relative reductions in diabetes incidence of 27% for lifestyle (hazard ratio 0.73, 95% CI 0.65–0.83) and 18% for metformin (hazard ratio 0.82, 95% CI 0.72–0.93) versus placebo. By year 15, more than half of participants in every group had developed diabetes, and the authors note that differences between groups declined over time.
21 years (Lancet Diabetes & Endocrinology, 2025): among 3,195 participants, the hazard ratios versus placebo were 0.76 (95% CI 0.68–0.85) for lifestyle and 0.83 (95% CI 0.74–0.93) for metformin. The rate differences were 1.59 fewer cases per 100 person-years for lifestyle and 1.17 fewer for metformin. Median diabetes-free survival was 3.5 years longer with lifestyle and 2.5 years longer with metformin. Hazard ratios are relative measures.

The ADA’s Section 3 adds that major cardiovascular events were not significantly different between groups at 21 years.

What do these numbers not tell you?

Thresholds are conventions. A result just under or just over a line isn’t a different kind of person; the ADA describes risk as continuous.
Prevalence figures are estimates. They are CDC estimates, not a head count of individual people.
The DPP enrolled a selected group with intensive supervision. Trial results are averages and don’t predict any one person’s course.
Groups converged over time. In follow-up, the gap between groups narrowed, and by 15 years more than half of participants in each group had developed diabetes.
The outcome was diabetes incidence, not heart or kidney outcomes. Glucose markers aren’t proof of future health events.
Prediabetes often travels with other risk factors, such as blood pressure and blood fats, according to the ADA.

What questions can you bring to your care team?

If you’re curious about your own results, these questions can help:

Which test was used, and was it repeated?
Where do my results sit within the range?
How often should they be rechecked?
Which other risk factors, such as blood pressure and blood fats, are worth checking?

A clinician who knows your history is the right person to interpret your results. For more explainers, see more Read explainers.

Frequently asked questions

What A1C level is prediabetes?

Under the ADA’s 2026 Standards, an A1C of 5.7–6.4% is in the prediabetes range, and 6.5% or higher is in the diabetes range. The ADA notes that risk is continuous across these values.

What fasting glucose level is prediabetes?

The ADA’s 2026 range is 100–125 mg/dL (5.6–6.9 mmol/L). The WHO and some other groups use a lower limit of 110 mg/dL.

How many Americans have prediabetes?

CDC estimates 115.2 million U.S. adults aged 18 and older have prediabetes (report page updated Sept 2026).

Does prediabetes always turn into type 2 diabetes?

No. In the DPPOS 15-year follow-up, cumulative diabetes incidence was 62% with placebo, 56% with metformin, and 55% with lifestyle, so most but not all participants in this high-risk trial cohort developed diabetes. Results in other groups differ.

Can lifestyle changes lower the chance of type 2 diabetes?

In the DPP trial, diabetes incidence was 4.8 cases per 100 person-years in the lifestyle group versus 11.0 with placebo over an average of 2.8 years. The ADA describes similar lifestyle goals. Any plan is a decision for a person and their clinician.

Sources

American Diabetes Association Professional Practice Committee. 2. Diagnosis and Classification of Diabetes: Standards of Care in Diabetes—2026. Diabetes Care 2026;49(Suppl 1):S27. (PMC)
American Diabetes Association Professional Practice Committee. 3. Prevention or Delay of Diabetes and Associated Comorbidities: Standards of Care in Diabetes—2026. Diabetes Care 2026;49(Suppl 1):S50. (PMC)
Centers for Disease Control and Prevention. National Diabetes Statistics Report. Updated Sept 2026.
Diabetes Prevention Program Research Group (Knowler WC, et al.). Reduction in the Incidence of Type 2 Diabetes with Lifestyle Intervention or Metformin. N Engl J Med 2002;346(6):393–403. (PubMed) Supported by NIDDK and other NIH institutes and U.S. agencies, the American Diabetes Association, Bristol-Myers Squibb, and Parke-Davis; Lipha Pharmaceuticals supplied the metformin and placebo. One author reported owning stock in Bristol-Myers Squibb.
Diabetes Prevention Program Research Group. Long-term effects of lifestyle intervention or metformin on diabetes development and microvascular complications over 15-year follow-up: the Diabetes Prevention Program Outcomes Study. Lancet Diabetes Endocrinol 2015;3(11):866–875. (PMC) Funding: NIDDK.
Knowler WC, et al. Long-term effects and effect heterogeneity of lifestyle and metformin interventions on type 2 diabetes incidence over 21 years in the US Diabetes Prevention Program randomised clinical trial. Lancet Diabetes Endocrinol 2025;13(6):469–481. (PMC) Funding: NIDDK.

This article is for general education. It isn’t a substitute for personal medical advice. If you have questions about prediabetes, blood sugar, or your lab results, talk with a licensed clinician who knows your health history.

Medical disclaimer: Content on Metabolic Media is for educational purposes only and is not medical advice, diagnosis, or treatment. We are not a medical provider, pharmacy, or telehealth clinic. Always consult a licensed clinician. If you have a medical emergency, call 911 (US) or your local emergency number.