Frontiers in Nutrition · Oct 30, 2026
Circadian-related rhythm robustness is associated with insulin sensitivity, insulin resistance, and triglycerides in healthy adults
By María Rodríguez-Martín, Ivan Dozzani, Jesualdo Tomás Fernández-Breis et al.
Circadian disruption has been linked to metabolic dysfunction, but the relationship between free-living circadian-related rhythm characteristics and metabolic regulation remains incompletely understood. We investigated cross-sectional associations between rhythm characteristics derived from wrist temperature, activity, and position (TAP) and glucose and insulin regulation in 716 adults (71% women; age 38 ± 14 years; BMI 25.75 kg/m²) from the ONTIME-MT study. TAP rhythms were continuously monitored for 7 days using wearables, and principal component analysis was used to derive TAP rhythmicity factors. Glucose and insulin regulation was assessed using a 2-hour oral glucose tolerance test, including glucose and insulin area under the curve (AUC), insulin sensitivity index, corrected insulin response, disposition index, and homeostatic model assessment of insulin resistance (HOMA-IR). Two TAP-derived factors explained 70.7% of the total variance. Factor 1 (52.8%) reflected rhythm robustness, whereas Factor 2 reflected rhythm timing (17.9%). Higher Factor 1 scores were associated with lower insulin AUC, fasting insulin, HOMA-IR, and triglycerides, and greater insulin sensitivity (all FDR < 0.05), with less consistent associations observed for glucose AUC and disposition index across sensitivity analyses. Higher Factor 2 scores, reflecting later TAP timing, were associated with higher fasting glucose in the main analysis. Factor 2 was also associated with later dim-light melatonin onset (P = 0.020), a marker of endogenous circadian phase. Exploratory mediation analyses suggested that triglycerides partially accounted for the associations between TAP rhythm robustness with insulin sensitivity (13%) and HOMA-IR (18%). Less robust TAP rhythms under free-living conditions were associated with a less favorable metabolic profile, particularly lower insulin sensitivity, greater insulin resistance and higher triglycerides. These findings support further investigation of wearable-derived circadian-related rhythm characteristics as non-invasive markers of metabolic health.