Related Experiment Video
Updated: Jul 19, 2026
Experimental Protocol of a Three-minute, All-out Arm Crank Exercise Test in Spinal-cord Injured and Able-bodied Individuals
Published on: June 8, 2017
High Test-Retest Reliability but Questionable Validity: Assessing a Step-Ramp-Step Protocol to Estimate Critical
Quentin Rousseau1,2,3, Giuseppe Rabita1, Emmanuel Brunet2
1French Institute of Sport (INSEP), Laboratory "Sport, Expertise and Performance" (EA 7370), Paris, FRANCE.
Purpose:
This study aimed to assess (1) the concurrent validity between critical power (CP) and the corrected power output at the respiratory compensation point (PO RCP ) and between W' and the quantity of energy depleted above PO RCP ( W'mod-SRS ) in a modified Step-Ramp-Step (mod-SRS) protocol; (2) the test-retest reliability of endurance variables derived from a mod-SRS protocol.
Methods:
Twelve trained male triathletes (31 ± 6 yr, 67.9 ± 7.1 mL min -1 kg -1 ) performed (1) an SRS protocol, (2) 4-5 severe-intensity constant work rate trials, (3) two mod-SRS protocols (test/retest), and (4) a heavy-intensity constant work rate trial. CP and W' were computed using a "best individual fit" approach. PO RCP was corrected appropriately, while W'mod-SRS was computed as the energy depleted above PO RCP . Test-retest reliability was assessed with the intraclass correlation coefficient (ICC), the standard error of measurement (SEM), and the coefficient of variation (CV%).
Results:
CP was higher than PO RCP in both trials (bias = 19 ± 19 W and 17 ± 20 W, P = 0.019 and 0.045). W' and W'mod-SRS were similar during both trials (bias = -2.8 ± 7.9 and -5.4 ± 8.8 kJ, P = 0.735 and 0.173). Test-retest reliability was moderate for W'mod-SRS (ICC = 0.66, SEM = 3.0 kJ, CV = 11.6%) and high for , gas exchange threshold (GET), RCP, and their associated PO values (ICCs = 0.85-0.95, SEMs = 68-184 mL min -1 , and 7-8 W, CVs = 2.1%-3.7%).
Conclusions:
The mod-SRS protocol reliably measures , gas exchange threshold, and RCP. Corrected PO RCP systematically underestimated CP, which consequently impacts the computation of W' using a mod-SRS protocol. The design of the protocol may allow a more complete depletion of energy available above RCP while still eliciting .
