Fewer hours of sleep or lower-quality sleep may be associated with poorer blood glucose control among African-Americans with diabetes, according to report in the September 18 issue of Archives of Internal Medicine.

Kristen L. Knutson, Ph.D., of the University of Chicago, and colleagues conducted a study of 161 African-Americans with type 2 diabetes, including 42 men and 119 women who had an average age of 57.3.

During a 30- to 45-minute interview, participants answered questions about sleep quality and quantity, their diabetes and any complications, and whether or not they took insulin. Their waist-hip ratio was measured, and the patients reported their height and weight, from which body mass index (BMI) was calculated.

Each individual was given a sleep quality score that ranged from zero to 21, with scores of greater than five indicating poor sleep quality. In addition, the researchers calculated a perceived sleep debt for each individual, described as the difference between the number of hours of sleep participants said they got on a typical weekday and the amount of sleep they said they wanted to get.

Hemoglobin A1c (HbA1c) levels, which measure the control of blood glucose over time, were obtained from patients' medical charts. A level of 7 percent or lower is the recommended optimal level for HbA1c.

The individuals in the study reported that they slept an average of six hours per night -- 22 percent averaged at least seven hours and only 6 percent at least eight hours. About 71 percent had sleep quality ratings of greater than five, indicating poor sleep quality. The average HbA1c level was 8.3 percent and only 26 percent had a level below the recommended 7 percent.

Thirty-nine patients, who reported that their sleep was frequently disrupted by pain, were excluded from further analyses.

Among the remaining 35 men and 87 women, 67 percent had poor sleep quality. Higher HbA1c levels were associated with lower sleep quality, less sleep and a larger perceived sleep debt, even after researchers controlled for sex, age, BMI, complications and use of insulin.

Participants were then classified based on whether or not they had complications from diabetes and whether or not they used insulin.

"In patients without complications, perceived sleep debt but not subjective sleep quality was associated with lnHbA1c levels," the researchers reported. "In contrast, in patients with at least one complication, [sleep quality] score, but not perceived sleep debt, was a significant predictor after controlling for covariates."

The results do not indicate whether diabetes control impacts sleep -- for instance, excessive urination at night resulting from high blood glucose levels could interrupt sleep -- or whether insufficient or poor-quality sleep could contribute to poor glucose control.

"Additional research is needed to determine whether optimizing sleep duration and quality may improve glucose control in patients with type 2 diabetes," the researchers concluded.