Energy expenditure during level human walking: seeking a simple and accurate predictive solution

Area of Science:

• Exercise Physiology
• Biomechanics
• Metabolic Research

Background:

• Accurate prediction of walking metabolic energy is crucial for health, physical status, and fitness evaluations.
• Existing equations like ACSM and Pandolf et al. show limitations in predicting human walking metabolism across diverse populations.
• Understanding the relationship between walking speed, height, and metabolic cost is essential for developing precise predictive models.

Purpose of the Study:

• To identify a concise, generalized equation for predicting human walking metabolism.
• To compare the predictive accuracy of existing literature equations (ACSM, Pandolf et al., HWS) against aggregated literature data.
• To evaluate the goodness-of-fit of one-component versus two-component models for describing walking metabolism.

Main Methods:

• Aggregated 127 literature metabolic rate values (speed range: 0.4–1.9 m/s) from 25 diverse subject populations.
• Calculated population-specific resting metabolic rates (V̇o2 rest) using standardized equations.
• Compared predictive accuracy (SEE) of ACSM, Pandolf et al., and HWS equations; derived and compared one- vs. two-component models (linear, exponential, exponential/height).

Main Results:

• ACSM and Pandolf et al. equations significantly underpredicted observed metabolic rates, yielding higher standard errors of estimate (SEE ≈ 4.4) compared to HWS (SEE = 1.13 ml O2·kg(-1)·min(-1)).
• Two-component models demonstrated substantially better goodness-of-fit (R² = 0.90 ± 0.03) and lower predictive errors (SEE = 1.21 ml O2·kg(-1)·min(-1)) than one-component models (R² = 0.63 ± 0.1, SEE = 2.22 ml O2·kg(-1)·min(-1)).
• A new generalized equation, V̇o2 total = V̇o2 rest + 3.85 + 5.97·V(2)/Ht, was identified as a concise predictor of walking metabolism.

Conclusions:

• The HWS equation and two-component models offer superior accuracy for predicting human walking metabolism compared to ACSM and Pandolf et al. equations and one-component models.
• The newly derived generalized equation provides a highly accurate and concise method for estimating walking metabolic energy expenditure.
• This improved predictive model has significant implications for applications in health monitoring, fitness assessment, and understanding human physical performance.