Diagnosis of diabetes mellitus requires evaluation of plasma glucose measured in the morning after an eight-hour fast. Troisi and associates conducted an analysis of the National Health and Nutrition Examination Survey to determine if an afternoon fasting measurement of plasma glucose can be used for the diagnosis of diabetes mellitus.

Researchers collected demographic information about the patients, including whether their phlebotomy was performed in the morning or afternoon. Each participant was instructed to fast (after 8:30 p.m. if venipuncture was performed in the morning and after 11:30 a.m. if it occurred in the afternoon). Those with a previous diagnosis of diabetes mellitus and those who were pregnant were excluded. Undiagnosed diabetes mellitus was defined as fasting plasma glucose levels of 126 mgper dL (7.0 mmol per L) or more; impaired fasting plasma glucose was defined as 110 to 125 mg per dL (6.1 to 6.9 mmol per L).

Participants in the morning and afternoon groups were similar in all regards except glucose levels. Those in the morning group had higher mean fasting plasma glucose levels than those in the afternoon group (97.4 mg per dL [5.4 mmol per L] compared with 92.4 mg per dL [5.1 mmol per L]). Fasting levels were highest in the early morning and gradually declined as the day went on, with a trend that was statistically significant. The length of the fast was not an important factor. Morning glucose readings were higher than afternoon readings regardless of the participant's age. The difference in the fasting plasma glucose level was enough to indicate that the prevalence of diabetes mellitus in the morning group was twice the prevalence in the afternoon group (2.8 percent versus 1.4 percent). Impaired fasting glucose also was more prevalent in the morning group (7.4 percent) versus the afternoon group (2.6 percent).

The authors conclude that the time of day may have a clinically significant impact on fasting plasma glucose levels, and that confirmatory testing on a different day should be performed in patients being tested for diabetes mellitus. If the original test is performed in the afternoon, a follow-up morning test is recommended.