Both time-to-exhaustion (TTE) and time-trial (TT) exercise tests are commonly used to assess exercise performance, but no study has directly examined the reliability of comparable tests in the same subjects.
Purpose: To evaluate the reliability of comparable TTE and TT treadmill running tests of high and moderately high exercise intensity in endurance-trained male distance runners, and to validate Hinckson and Hopkins TT prediction methods using log-log modeling from TTE results.
Methods: After familiarization tests, eight endurance-trained male distance runners performed, in a randomized, counterbalanced order, eight trials consisting of two 5-km TT and two 1500-m TT, and four TTE tests run at a speed equivalent to the average speed attained during both the 5-km and 1500-m TT distances.
Results: Typical error of the estimate (TEE) expressed as a coefficient of variation for the 5-km TT, 5-km TTE, 1500-m TT, and 1500-m TTE were 2.0, 15.1, 3.3, and 13.2%, respectively. The standard error of the estimate for predicted TT running speed using log-log modeling from TTE results was 0.67%, and the predicted versus criterion reliability of this method revealed TEE values of 1.6% and 2.5% for the prediction of 5-km and 1500-m TT, respectively.
Conclusion: The variability of 5-km and 1500-m TT tests was significantly less than for similar TTE treadmill protocols. Despite the greater variability of the TTE tests, log-log modeling using the TTE test results reliably predicted actual TT performance.