Measuring glycated hemoglobin and fasting glucose from the same blood sample can confirm diabetes, a study in the Annals of Internal Medicine suggests.

Researchers followed roughly 1,000 patients who had elevated measures of glycated hemoglobin or fasting glucose (or both) at baseline in single blood samples in 1990. None of the patients received an immediate diabetes diagnosis. (Roughly one third had elevations in both measures, and two thirds had elevations in just one.)

Diabetes diagnoses were established during follow-up via self-report or physician diagnosis over a 15-year period. Those with elevations in both measures at baseline had a hazard ratio for diabetes over follow-up of 25.0, compared with nondiabetics; for those with just one elevated measure, the ratio was 5.75.

By the 15-year mark, the positive predictive value for the two-elevations-in-one-sample diagnosis definition was almost 90%.

Editorialists say the findings need replication "before becoming accepted clinical practice."