Relationships Between Sleep and Self-Care in Type 2 Diabetes: An Ecological Momentary Perspective

Aim: Sleep disturbance is a widespread yet often ignored complaint among people with type 2 diabetes. Sleep disturbance has been an under-examined risk factor for poor diabetes self-care. The aim of this study was to examine whether sleep disturbance is related to diabetes self-care using longitudinal data in the real-world setting. Methods: A correlational, longitudinal design was used. A convenience sample of 64 older adults with type 2 diabetes was recruited. Validated instruments were used to measure subjective sleep quality, self-care, and other primary constructs including self-efficacy, distress, fatigue, and daytime sleepiness. Participants completed a sleep and self-care diary each day for eight days to assess their subjective sleep and self-care behaviors (e.g., eating behavior, physical activity, and medication adherence). The ActiGraph was used to measure daily objective sleep and physical activity. Data were analyzed using multiple linear regression analyses and mixed-effect models. Results: Regression analysis suggested that six variables explained 51% of the variation in overall self-care. Subjective sleep quality was a strong predictor of overall self-care. The effect of sleep quality (β=-0.26) on self-care was smaller than that of the commonly reported diabetes distress (β=-0.39) but larger than that of daytime sleepiness (β=-0.21). Mixed-effect models indicated that total sleep time was negatively related to light-intensity physical activity the following day (estimates: -0.24 to -0.43). Based on the mixed-effect models, although sleep alone did not predict eating behavior the following day, its interaction with morning fatigue was a significant predictor. These findings suggested that the effect of sleep on eating behavior (e.g., conscious restriction on eating, loss of control over eating, and eating in response of emotional cues) was different for people with different levels of fatigue. Sleep was not associated with medication adherence. Conclusions: In older adults with type 2 diabetes, poor sleep quality is associated with poor self-care. Sleep likely affects self-care behaviors, especially eating behavior, through its interaction with fatigue. There is little evidence supporting the significant relationships between night sleep and daytime accelerometer-derived physical activity and medication adherence.