Extended Data Fig. 3: Changes in VAT volume (cm3; A and F), VAT percentage (%; B and G), body weight (kg; C and H), fat–free mass (kg; D and I), and fat mass (kg; E and J) in men and women among the UC, early TRE, late TRE, and self–selected TRE groups after the 12-week intervention.
Data are raw means with 95% confidence interval. Similar letters represent significant differences between groups as determined by post-hoc Bonferroni correction for multiple comparisons (P < 0.05). The P values reported are two-sided and Bonferroni-adjusted P values. Secondary analysis (body weight changes in kilograms) between early time-restricted eating (TRE) and usual care (UC, men: P < 0.008; women: P = 0.003), late TRE and UC (women: P = 0.005), and self-selected TRE and UC (men: P < 0.001). Changes were calculated as the difference between postintervention and preintervention values. VAT percentage indicates changes relative to preintervention values. Intervention effects on primary and secondary outcomes at 3 months after the intervention were assessed based on repeated-measures linear mixed-effects multilevel models, which included random cluster (site) effects. Individual measures of change were therefore modelled as the function of randomly assigned group, site, assessment time, and their interaction terms, stratifying the analysis by sex (that is, separately for men and women). Model-based estimations were performed with an intention-to-treat approach using the restricted maximum-likelihood method, the model assuming that missing values were missing-at-random.
