This repeated‑measures crossover study evaluated whether a 1% treadmill grade reproduces the energetic demands of overground running in trained endurance athletes. Twelve nationally licensed male runners (mean age 21.3 ± 2.4 years) completed a ramp test to determine V̇O2max and two identical stepwise protocols (8–15 km·h⁻¹; 3‑min stages) performed indoors on a treadmill and outdoors on a 400 m track. Breath‑by‑breath gas exchange was averaged over the final 60–90 s of each stage to obtain steady‑state V̇O2. Primary outcomes were running economy (RE), oxygen cost of transport (O2‑COT), and energy cost (EC). Repeated‑measures ANOVA showed significant main effects of Condition and Intensity for all three outcomes (all p ≤ 0.012). Overground running yielded higher RE (51.3 ± 2.2 vs. 49.6 ± 3.1 mL·kg⁻¹·min⁻¹; p‑FDR = 0.01), higher O2‑COT across intensities (204.2 ± 11.3 vs. 197.1 ± 12.3 mL·kg⁻¹·km⁻¹; p‑FDR ≤ 0.01), and consistently greater EC (3.87–4.60 vs. 3.65–4.10 J·kg⁻¹·m⁻¹; p‑FDR < 0.001). Carbohydrate oxidation increased during overground running at intensities approaching VT2 (p‑FDR = 0.02). The authors conclude that the 1% treadmill grade does not fully replicate overground energetic demands, particularly near the second ventilatory threshold.
Endurance running requires coordinated cardiorespiratory and metabolic regulation across exercise intensities. Measures describing energetic cost include running economy (RE) (steady‑state oxygen uptake normalized to body mass), oxygen cost of transport (O2‑COT) (oxygen per distance), and energy cost (EC) (joules per mass per distance integrating substrate oxidation). Physiological intensity domains are commonly defined by ventilatory thresholds (VT₁ and VT2), which provide a relative framework for comparing metabolic responses between conditions.
A 1% treadmill grade has been recommended historically to approximate overground energetics during laboratory testing. However, prior meta‑analysis and mechanistic work have questioned whether a universal 1% correction is physiologically valid across speeds and populations. The present study addresses gaps in previous work by comparing distance‑normalized outcomes (O2‑COT and EC) alongside RE, anchoring comparisons to individual ventilatory thresholds, and reporting substrate oxidation patterns.
Twelve male endurance athletes (Tier 3: highly trained/national‑level) completed testing after institutional ethics approval and informed consent. Participants trained ≥5 sessions per week with weekly volumes of about 65–80 km and were free from cardiopulmonary, metabolic, or musculoskeletal disorders. Testing occurred between 01/07/2025 and 01/08/2025.
Study design was a within‑subject crossover. A ramp test established V̇O2max and ventilatory thresholds. Two identical stepwise protocols (3‑min stages from 8 to 15 km·h⁻¹) were completed: one on a treadmill set to 1% grade indoors and one overground on a 400 m track outdoors. Breath‑by‑breath gas exchange was collected and averaged over the final 60–90 s of each stage to compute steady‑state V̇O2, substrate oxidation, and caloric equivalents. Outcomes included RE (mL·kg⁻¹·min⁻¹), O2‑COT (mL·kg⁻¹·km⁻¹), and EC (J·kg⁻¹·m⁻¹). Statistical analysis used two‑way repeated‑measures ANOVA (Condition × Intensity domain) with false discovery rate correction for multiple comparisons.
Significant main effects of Condition and Intensity domain were observed for RE, O2‑COT, and EC (all p ≤ 0.012). RE was higher during overground running: group means were 51.3 ± 2.2 mL·kg⁻¹·min⁻¹ (overground) versus 49.6 ± 3.1 mL·kg⁻¹·min⁻¹ (treadmill) with p‑FDR = 0.01. The divergence in RE emerged primarily at intensities between 70% and 100% of VT2.
O2‑COT was consistently greater during overground running across all intensity domains (204.2 ± 11.3 vs. 197.1 ± 12.3 mL·kg⁻¹·km⁻¹; p‑FDR ≤ 0.01). The absence of a significant Condition × Intensity interaction for O2‑COT indicates the between‑condition difference was stable across the examined intensity range.
Energy cost (EC) also remained higher during overground running across intensities (reported ranges: 3.87 ± 0.21 to 4.60 ± 0.27 J·kg⁻¹·m⁻¹ overground vs. 3.65 ± 0.22 to 4.10 ± 0.18 J·kg⁻¹·m⁻¹ treadmill) with p‑FDR < 0.001. A significant Condition × Intensity interaction was present for EC (p = 0.045), indicating the magnitude of the difference varied with intensity.
Substrate oxidation analysis showed significantly greater carbohydrate oxidation during overground running at intensities approaching VT2 (p‑FDR = 0.02), implying a higher glycolytic contribution near higher submaximal workloads.
The study demonstrates that laboratory treadmill testing using a 1% grade underestimates energetic demand relative to overground running in trained endurance athletes. Differences were observed across outcome metrics: RE showed divergence primarily near high‑intensity domains, O2‑COT was consistently greater overground across intensities, and EC differences varied by intensity. Increased carbohydrate oxidation near VT2 during overground running provides a metabolic explanation for elevated EC and RE at higher relative intensities.
These results challenge the assumption that a single 1% grade correction uniformly matches overground energetics and highlight the value of distance‑normalized measures (O2‑COT and EC) and threshold‑anchored comparisons when translating laboratory findings to field performance.
In this cohort of highly trained male endurance runners, overground running imposed higher oxygen and energy costs than treadmill running at 1% grade, particularly near VT2. The commonly applied 1% treadmill grade correction does not fully replicate overground energetic demands, which has implications for CPET‑based performance assessment and interpretation. Future work should examine broader populations, different speeds, and environmental conditions to refine laboratory‑to‑field translations.