Background: Sleep has been identified as having an influence on the success of weight-loss interventions; however, knowledge of the mechanisms and the extent to which sleep disturbances affect the magnitude of weight reduction is inconclusive.
Objective: To determine if sleep duration and quality can predict the magnitude of weight reduction in a weight-loss intervention program for overweight and obese women.
Methods: Ninety overweight and obese women aged 25–65 years completed the 7-month weight-loss phase of our weight-loss intervention. Sleep duration and quality were evaluated before the intervention by the Pittsburg Sleep Quality Index (PSQI), a self-report questionnaire, and by actigraphy. Serum levels of ghrelin, leptin, cortisol and insulin also were measured at baseline. Insulin resistance was measured by the homeostasis model assessment of insulin resistance (HOMA-IR).
Results: The mean reduction rate of body mass index (BMI) after the intervention was 13.6%. Multiple linear regression revealed that the number of wake episodes (WEs) per night had a significant relationship with the reduction of BMI even after adjusting for other clinical variables (β=−0.341, P=0.001). The participants with five or more WEs per night (high-WE group) had a significantly lower reduction in BMI compared with those with fewer than five (normal-WE group), after adjusting for confounding variables. In contrast, the PSQI-assessed parameters, reflecting the subjective assessments of sleep quality and duration, failed to detect an association with the reduction in BMI. Baseline HOMA-IR was significantly higher in the high-WE group than in the normal-WE group after adjusting for confounding variables.
Conclusions: Higher sleep fragmentation, as manifested by the increased number of WEs, predicts a lower magnitude of weight reduction in persons participating in weight-loss programs.
Alex’s Notes: A recent meta-analysis of over 600,000 adults reported a 55% increased risk of obesity in those who had fewer than five hours of sleep, with a decrease in BMI by 0.35 kg/m2 for each additional hour of sleep. Sleep matters. But the time we spend in bed is not always sleep.The aim of this study was to determine if objective and subjective sleep parameters predict the magnitude of weight loss in overweight and obese women on a weight-loss intervention program.
The study consisted of two phases. The first was a weight-loss phase that provided intensive behavioral therapy over seven months. If the participants lost at least 5% of their bodyweight, then they were moved to phase 2, which was a 3-month weight maintenance phase with a follow-up of 34 months. All the participants were women aged 20-65 years and a BMI of at least 25 kg/m2. None had eating disorders, were on any bodyweight affecting medications, or had sleep apnea.
The weight-loss phase
The treatment program was conducted in small groups of ten persons, with each person participating in 35 group sessions and five individual sessions over the 44-week period. Each session lasted 90 minutes and the sessions were conducted once a week for the first consecutive 34 sessions, and the remaining sessions were held once every 2 weeks.
“All participants were instructed to keep a daily food diary to track their consumption of all food and drinks, wear a pedometer and record their daily number of steps. They also were advised to reduce their dietary intake by 500 calories per day from their caloric intake at the beginning of their therapy. The nutritionists checked the nutritional balance of the participants’ diets by examining their recordings of their food intake, and advised participants about the importance of eating vegetables and reducing the consumption of fat foods and sweets. The participants were advised to increase their level of physical exercise to a moderate intensity, such as walking (8000–10 000 steps per day). Furthermore, a series of stress-management sessions also was included in the program.”
Body composition was measure with DXA, depression was evaluated using the Japanese version of the Center for Epidemiologic Studies-Depression Scale (CES-D), and objective sleep parameters were obtained via 5-7 consecutive days of actigraphy. Subjective sleep parameters also were evaluated by the Pittsburgh Sleep Quality Index (PSQI).
Overall, 119 women participated but only 90 lost at least 5% of their bodyweight, with the average weight lost being nearly 15% the initial body weight. The average age was 48 years, and the average initial body fat was 38%. Amusingly, despite weighing 77.6kg (170 lbs) at baseline and being clinically obese, the reported baseline calorie intake was 1786 kcal per day, which was reduced to 1176 kcal per day after the weight loss phase. All according to self-report food logs of course. Steps also increased from 6400 to 8000 per day.
But we are interested in sleep
Yes, clearly the weight loss phase was a huge success for most of the women, but how did sleep influence the outcomes? At baseline, the average sleep duration was just under six hours, and the average number of wake episodes was 5.72 per night. The authors point out that the average number of waking episodes in normal sleepers is reported to be four. A significant association (albeit modest magnitude) was found between the reduction in BMI and body fat percentage and the baseline sleep efficiency and waking episodes. Of all the independent variables, the number of waking episodes was the strongest predictor of BMI reduction, and this remained significant even after adjustment for age, baseline BMI, smoking status, CES-D score, average daily caloric intake, and average number of daily steps during treatment.
In addition to the above, the baseline hormonal markers showed that homeostasis model assessment of insulin resistance (HOMA-IR) and cortisol levels were significantly different between the normal- and the high-wake episode groups, with none of the other baseline hormonal markers differing.
What does this mean?
It means that further studies are needed to clarify whether short sleep duration per se or short sleep due to frequent interruptions causes an increased risk of obesity and reduced weight loss.
More interesting in my opinion is the interaction with the hormonal differences. For instance, one study elicited fragmented sleep across all of the sleep stages in 11 healthy volunteers for two nights in a laboratory setting, using auditory and mechanical stimuli. They showed that sleep fragmentation decreased both insulin sensitivity and glucose effectiveness. The HOMA-IR was significantly higher in the high-wake episode group of the current study, suggesting that fragmented sleep may indeed attenuate weight loss by inducing insulin resistance.
As for cortisol, it is normally highest in the morning and yet was shown to be lower in the high-wake episode group of the current study, suggesting that sleep fragmentation may also cause dysregulation of the hypothalamic-pituitary-adrenal axis. No doubt this would attenuate weight loss as well. But still, the results are only associations and they cannot be extended to men.