Fig. 3: Personal profiling of meal timing-related dietary habits and their associations with glucose metabolic outcomes.

a Violin plots showing timing-related diet features in the cohort. The violin plots illustrate kernel probability density of the data at different values and the horizontal bar depicts the median of the distribution. The error bars represent the data within 1.5 times the IQR from the lower and upper quartiles. First food time (am), time of eating the first food of the day; last food time (pm), time of eating the last food of the day; daily eating span (sec), eating time window between the first food and the last food; last food ~ bed time (sec), time spent from the last food till the bed time; wake-up time ~ first food (sec), time spent from wake-up in the morning till eating the first food. Sleep-related diet parameters were derived through time-matching. b Forest plot showing individual associations of diet parameters with glucose metabolic outcomes using LASSO feature selection combined with multiple linear regression. A horizontal panel in the plot represents each glucose outcome model (i.e., metrics comprising metabolic tests and CGM). Associations that achieved statistical significance (BH-adjusted P < 0.1) between diet parameters and glucose outcomes are listed in this figure. The coefficient of each diet feature (a point of estimate depicted as the central marker) was derived from the covariate-adjusted multiple linear regression models (all diet features, age, sex, BMI, and ethnicity). The error bars represent the 95% confidence interval for the point estimate. %E Meal, energy proportion (%) of the meal timing to the total daily energy intake; %Carb, carbohydrate proportion (%) of the food group out of the total daily carbohydrate intake from all food groups; FPG fasting plasma glucose, IR insulin resistant, SSPG steady-state plasma glucose, representing muscle insulin resistance. Hyperglycemic range for 24 h was defined as >140 mg/dL, and for night time as >100 mg/dL. Time in target range for 24 h was defined as 70–140 mg/dL and for night time as 70–100 mg/dL.