Randomized Controlled Trial of Lifestyle Plus Sleep Extension Intervention on Adults With Prediabetes and Short Sleep

Background: Short sleep duration is a modifiable barrier to diabetes prevention, yet no adequately powered trials have tested whether integrating a behavioral sleep intervention into a lifestyle intervention improves glycemic or weight loss outcomes in adults with prediabetes. Objective: To compare the effects of a lifestyle versus a lifestyle plus sleep intervention on glucose control and weight loss in adults with prediabetes and habitual short sleep duration. Methods: In this two-arm, parallel-group randomized controlled trial, adults with A1c-confirmed prediabetes and actigraphy-confirmed short sleep duration completed two clinic visits and 2-week at-home monitoring periods, including actigraphy-estimated sleep and continuous glucose monitoring. The 8-week intervention was delivered one-on-one via a Health Insurance Portability and Accountability Act-compliant videoconferencing platform. The primary outcome was change in the percentage of time sensor glucose was above tight range (> 140 mg/dL). Secondary outcomes included actigraphy-estimated sleep parameters, self-reported sleep quality and daytime sleepiness, and body weight. We compared pre- to post-intervention changes between groups using a modified intention-to-treat approach. Results: Among 196 participants, we found no significant between-group difference in the primary glucose outcome. Both groups achieved significant within-group increases in sleep and reductions in body weight, with no significant between-group differences for either outcome. The lifestyle plus sleep group showed significantly greater improvements in self-reported sleep quality and daytime sleepiness compared to the lifestyle group. Sleep onset advanced significantly only in lifestyle plus sleep group, though the between-group difference did not reach statistical significance. A post hoc analysis found that earlier sleep onset, regardless of group assignment, was independently associated with greater weight loss, whereas changes in sleep duration were not. Discussion: Adding a behavioral sleep intervention to a lifestyle intervention did not improve glucose control. However, advancing sleep onset time predicted greater weight loss, independent of sleep duration or group assignment, identifying circadian timing as a promising and modifiable target for diabetes prevention. Prospective trials powered to test sleep timing interventions on metabolic outcomes are warranted.