0133 The Impact of Stress and Sleep: Capturing Multiday Patterns

Abstract Introduction Students commonly experience stress due to constant academic demands and societal pressures. This study aims to investigate whether there are differences in sleep efficiency among students experiencing high, low, and medium levels of stress. We performed statistical analyses on sleep efficiency across various stress levels to investigate how stress affects daily sleep patterns. Methods Using the GLOBEM dataset, we computed the range of daily sleep efficiency scores for individual students, considering their highest and lowest sleep efficiency levels observed throughout the study. We classified days into high, medium, and low-stress categories based on students’ weekly Perceived Stress Scale 4 (PSS-4) survey scores. We focused on sleep efficiency ranges within each stress group, emphasizing the significance of considering variability alongside average values to uncover distinct sleep patterns. A one-way analysis of variance (ANOVA) was employed to examine variations in sleep efficiency ranges among different stress groups. Subsequently, Tukey's post hoc analysis was conducted to identify specific groups that exhibited differences in their sleep efficiency. Results No statistically significant differences were observed in daily sleep efficiency scores among students with high, medium, and low stress levels. However, we identified significant variations in sleep efficiency ranges (ANOVA F(2) = 18.14, p < 0.0000001). Post-hoc Tukey analysis revealed mean differences, indicating a significant increase in sleep efficiency ranges between high and low-stress groups (p = 0.0) and a significant increase between high and medium-stress groups (p = 0.0). Notably, no statistically significant difference was found between the low and medium-stress groups. The instances of high-stressed days (N=140), medium-stressed days (N=2271), and low-stressed days (N=1615) provide context to these findings highlighting variations in sleep efficiency. Conclusion A significant correlation between stress severity and sleep efficiency does not exist at the day level. However, when we looked at variations in sleep efficiency over time, there are statistical significant differences that correlate with stress severity. The increase in sleep efficiency ranges among high-stress students indicates that they don’t consistently experience poor sleep efficiency but rather exhibit more erratic sleep patterns. This underscores the importance of considering stress severity when analyzing and addressing sleep patterns in students. Support (if any)