Beyond BMI: Clinical Features Supporting a True Type 2 Diabetes Diagnosis

In this portion of the discussion, Samson asks Umpierrez to walk through how the 2026 AACE classification algorithm helps confirm or question a diagnosis of type 2 diabetes in clinical practice. Umpierrez begins by reviewing that the biochemical thresholds used to diagnose diabetes—fasting plasma glucose of at least 126 mg/dL, 2-hour glucose of at least 200 mg/dL on oral glucose tolerance testing, hemoglobin A1c of at least 6.5%, or a random glucose of at least 200 mg/dL in the setting of hyperglycemic symptoms—apply to both type 1 and type 2 diabetes. These criteria, he notes, do not distinguish between etiologies but define the presence of diabetes itself.

To support a diagnosis of type 2 diabetes, Umpierrez outlines several clinical features that are commonly associated with this phenotype. These include overweight (body mass index [BMI] 25–29.9 kg/m²), obesity (BMI ≥30 kg/m²), a history of weight gain, prior gestational diabetes, a strong family history of type 2 diabetes, and physical examination findings of insulin resistance, such as acanthosis nigricans. He emphasizes acanthosis as one of the most important physical signs indicating significant insulin resistance. Samson and Umpierrez agree that these features can increase the likelihood of type 2 diabetes but must be interpreted within a broader clinical context.

However, both experts stress the limitations of relying on BMI alone. Umpierrez notes that BMI cutoffs are based on Western populations and may differ by race and ethnicity—for example, lower BMI thresholds in some Asian groups may better reflect cardiometabolic risk. He cites data from the adult-onset type 1 diabetes cohort discussed earlier, in which approximately 59% of patients with late-onset type 1 diabetes were overweight, and about 12% of adults with type 2 diabetes had a normal BMI. Samson and Umpierrez use these data to reinforce a key message of the 2026 AACE algorithm: although BMI and obesity are important clues, neither should be treated as definitive evidence of type 2 diabetes in adults with hyperglycemia.