Associations between Neurocardiovascular Signal Entropy and Physical Frailty

Figures

• 

  Figure A1. Results from robust multivariate linear regression showing beta coefficients (β) and 95% confidence intervals (CI) for both approximate entropy (ApEn) and sample entropy (SampEn) with the embedded dimension (m) set to 2 (as reported in the main manuscript), 3, and 4. All models adjusted for age, sex, education, diabetes, number of cardiovascular conditions, antihypertensive medication use, alcohol consumption habits, smoking, and depression. Model for active stand data additionally controlled for stand time. Abbreviations: sBP: systolic blood pressure; dBP: diastolic blood pressure; MAP: mean arterial pressure; HR: heart rate; O2Hb: oxygenated hemoglobin concentration; HHb: deoxygenated hemoglobin concentration; TSI: tissue saturation index.

• 

  Figure 1. Flow chart describing sample selection and exclusions.

• 

  Figure 2. Example plots of data from three different participants with “low” (0.20), “medium” (0.45), and “high” (0.70) levels of approximate entropy (ApEn) measured in the resting state systolic blood pressure (sBP) data. Individuals with higher entropy in one measure investigated (in this case, resting state sBP) also generally had higher entropy in the other physiological measures investigated, as visually illustrated above.

• 

  Figure 3. Results from robust multivariate linear regression showing beta coefficients (β) for both approximate entropy (ApEn) and sample entropy (SampEn). All models adjusted for age, sex, education, diabetes, number of cardiovascular conditions, antihypertensive medication use, alcohol consumption habits, smoking, and depression. Model for active stand data additionally controlled for stand time. Abbreviations: sBP: systolic blood pressure; dBP: diastolic blood pressure; MAP: mean arterial pressure; HR: heart rate; O2Hb: oxygenated hemoglobin concentration; HHb: deoxygenated hemoglobin concentration; TSI: tissue saturation index; β: beta coefficient.