Sleep assessment in aging adults with type 2 diabetes: agreement between actigraphy and sleep diaries

Actigraphy and sleep diaries have been widely used to evaluate various sleep parameters. However, their agreement in diabetes patients remains unclear. The objective of this study was to examine the agreement between sleep outcomes measured by actigraphy and sleep diaries in aging adults with type 2 diabetes (T2D).

A convenience sample of 53 T2D adults (aged 50-76 years) were enrolled. Participants wore a wrist ActiGraph and filled out a daily sleep diary for eight days. Total sleep time (TST), sleep efficiency (SE), sleep onset latency (SOL), and wake after sleep onset (WASO) were obtained from the actigraphy and sleep diaries. Bland-Altman plots were conducted to examine the agreement between each sleep outcome.

The differences for TST and SE assessed by actigraphy and sleep diaries were 11.3 min (SD 65.3) and 0.2% (SD 10.5). Bland-Altman plots revealed wide limits of agreement between actigraphy- and diary-measured TST (95%CI: -139.3 min, 116.7 min) and SE (95%CI: -20.9%, 20.4%). Systematic biases were present for WASO and SOL: compared to actigraphy, sleep diaries underestimated WASO and overestimated SOL. As the SOL and WASO increased, the agreement became lower.

Overall, the agreement between actigraphy and sleep diaries is poor across all measures in aging adults with T2D patients. Findings from this study highlight the need for sleep researchers and clinicians to consider the method used for sleep assessment when developing interventions or interpreting study findings.