Estimated eating-window phase versus eating jetlag in relation to insulin resistance among adults without diagnosed diabetes: A two-period NHANES analysis

Chrononutrition includes eating-window timing and weekday–weekend shifts in eating time. We analyzed the 2017–March 2020 and 2021–2023 National Health and Nutrition Examination Survey periods among adults aged ≥20 years without diagnosed diabetes who had two 24-hour dietary recalls (one Monday–Thursday and one Friday–Sunday) and fasting data. Homeostasis model assessment of insulin resistance (HOMA-IR) was the primary outcome. Estimated eating-window phase was the circular 4:3 weighted mean of weekday and weekend eating midpoints, defined by the first and last eating events containing ≥50 kcal; absolute eating jetlag was the shortest circular difference between these midpoints. Survey-weighted models included both exposures. Samples comprised 1,492 and 709 participants. Each 1-hour later estimated eating-window phase was associated with 6.72% higher HOMA-IR (95% CI, 2.45%–11.16%; P=0.003) in 2017–March 2020. The corresponding estimate was 2.98% (95% CI, −1.86% to 8.06%; P=0.214) in 2021–2023. Absolute eating-jetlag estimates were near null in both periods. The phase estimates did not differ significantly between periods (P=0.235). A cumulative-energy caloric-midpoint phase was not associated with HOMA-IR in either period. Eating-window phase, eating jetlag, and caloric timing may represent distinct behavioral dimensions, although estimates from two recalls warrant cautious interpretation.