Association between sleep duration and hemoglobin A1c level

Objective: The association between sleep habits and hemoglobin A(1c) (HbA(1c)) level has not been sufficiently examined. In the present study of residents in a local community, the associations between sleep duration and HbA(1c) level were examined. Methods: Self-administered questionnaires were mailed to 1062 residents in a rural community in Japan, and completed questionnaires were collected. At the time of collection, the fasting plasma glucose and HbA(1c) levels were measured using peripheral blood samples. For the analyses, values that were considered to represent high levels were a fasting plasma glucose level of >= 126 mg/dl and a HbA(1c) level of >= 6.5%. Logistic regression analyses were performed to examine the associations between sleep duration and high fasting plasma glucose or high HbA(1c), levels. Results: The prevalence of high fasting plasma glucose and high HbA(1c) levels was significantly high (p < 0.01) in subjects with a short or a long sleep duration. Logistic regression analyses demonstrated a significant association between high HbA(1c) level and sleep duration. The adjusted odds ratios for a high HbA(1c) level showed high values with regard to both short and long sleep durations. Conclusions: HbA(1c) level showed a U-shaped association with sleep duration. These results suggest that there may be an appropriate range of sleep duration in individuals with glucose tolerance disorders. It is expected that the present findings will contribute to the treatment and prevention of diabetes mellitus. (C) 2007 Elsevier B.V. All rights reserved.