0168 Consistency of Sleep Regularity Measurement from Subjective Recall vs Objective Measurement

Introduction Irregular day-to-day sleep timing is associated with increased cardiometabolic risk, including hypertension, obesity, and diabetes, all of which are strong predictors of mortality. Because different measurement approaches (i.e., recall vs. objective measures) with potential varying observation lengths (i.e., 1-week vs 2-weeks) could yield differing timing estimates, understanding the level of agreement between approaches is crucial for both research and clinical applications. Thus, we assessed the similarity between sleep timing regularity derived from recall and objective daily sleep diaries across 7- and 14-day tracking intervals. Methods Twenty-five participants (18 F) reported previous 7-day recall of sleep onset and offset times via an online survey, then completed daily sleep-wake diaries for up to 14 days. We calculated the Sleep Regularity Index (SRI), reflecting day-to-day timing consistency, and the standard deviation (SD) of sleep onset from recall, the first 7-days of tracking, and the full 14-days of tracking. Associations between recall and objective measures for SRI and SD were tested using Pearson correlations. Intraclass correlation coefficients (ICC) from a two-way mixed-effects model assessed individual consistency with strengths classified as slight (0.00-0.20), fair (0.21-0.40), moderate (0.41-0.60), substantial (0.61-0.80), and almost perfect (0.81-1.00). Results There was a positive association between recall and 7-day objective tracking for SRI (r=0.58; p< 0.01) and a non-significant trend for an association between recall and SD of sleep onset (r=0.36; p=0.07). The consistency between recall and 7-day objective tracking was “moderate” for SRI (ICC = 0.53) and “fair” for SD of sleep onset (ICC=0.36). When comparing recall with 14 days of tracking, associations were significant for both SRI (r=0.62; p< 0.01) and SD of sleep onset (r=0.65; p< 0.01), with ICC values of “moderate” consistency for SRI (ICC=0.54) and “substantial” consistency for SD sleep onset (ICC=0.62). Conclusion Understanding how sleep regularity is measured may be essential for accurately interpreting its associations with health. Regularity from subjective recall was moderately consistent with objectively measured timing at 7-days of tracking, with stronger agreement over 14 days of tracking. These findings may underscore methodological considerations for researchers and clinicians when objective measures are impractical and suggest recall-based estimates may better capture long-term patterns than short-term regularity. Support (if any)