Device‐Measured 24‐h Movement Behaviors and Arterial Intima‐Media Thickness in Young Adults

Figures

• 

  FIGURE 1. Heat map ternary plots illustrating model‐based estimates of aortic intima‐media thickness (IMT) values for different 24‐h movement behavior compositions within the range observed in the study sample. Estimated aortic IMT values were obtained from a sex‐ and education‐adjusted four‐part compositional regression model including sleep, sedentary behavior (SED), light physical activity (LPA), and moderate‐to‐vigorous physical activity (MVPA). The color scale reflects estimated aortic IMT (mm), with red indicating higher values relative to the sample mean (0.60 mm), yellow values close to the sample mean, and green indicating lower values relative to the sample mean. Each panel represents a conditional slice of the four‐part composition, with the omitted behavior fixed at a predefined reference value corresponding approximately to the sample mean composition. Panel (A) displays the relative proportions of SED, LPA, and MVPA, while sleep was held constant at 480 min. For illustration, point X corresponds to a three‐part subcomposition consisting of 69% SED, 20% LPA, and 11% MVPA. The displayed proportions are relative to the three behaviors shown in the ternary plot and therefore sum…

• 

  FIGURE 2. Heat map ternary plots illustrating model‐based estimates of carotid intima‐media thickness (IMT) for 24‐h movement behavior compositions within the range observed in the study sample. Estimated carotid IMT values were obtained from the same four‐part compositional regression model and visualized using the same conditional‐slice approach as described in Figure 1. The color scale reflects estimated carotid IMT (mm), with red indicating higher values relative to the sample mean (0.46 mm), yellow values close to the sample mean, and green indicating lower values relative to the sample mean. Panels (A–D) correspond to the same fixed reference values and movement‐behavior combinations as in Figure 1, with the omitted behavior held constant at the corresponding reference value in each panel.

• 

  FIGURE 3. Estimated cross‐sectional differences in aortic intima‐media thickness (IMT) associated with theoretical one‐to‐one reallocations between 24‐h movement behaviors (e.g., from SED to LPA) relative to the mean 24‐h movement behavior composition in the study population (8.0 h sleep, 12.8 h SED, 1.7 h LPA, and 1.4 h MVPA). Estimates were obtained from a sex‐ and education‐adjusted compositional linear regression model. Shaded areas indicate 95% confidence intervals around the estimated differences.

• 

  FIGURE 4. Estimated cross‐sectional differences in carotid intima‐media thickness (IMT) associated with theoretical one‐to‐one reallocations between 24‐h movement behaviors (e.g., from SED to LPA) relative to the mean 24‐h movement behavior composition in the study population (8.0 h sleep, 12.8 h SED, 1.7 h LPA, and 1.4 h MVPA). Estimates were obtained from a sex‐ and education‐adjusted compositional linear regression model. Shaded areas indicate 95% confidence intervals around the estimated differences.