This research has been conducted in compliance with all applicable Federal Regulations governing the protection of human subjects in research. No human subjects were directly involved in this research.
ii SUMMARY BackgroundCardiorespiratory fitness is important for health, work, and athletic performance. Laboratory tests of maximal oxygen uptake (VO 2max ) are the gold standard for assessing this aspect of fitness. VO 2max protocols with small measurement errors will provide the best estimates of relationships between fitness and its antecedents and consequences. For example, tests with smaller errors will provide better indications of how well running tests function as substitutes for laboratory tests.
ObjectivePublished studies of the reliability of VO 2max tests provide an empirical basis for estimating VO 2max test precision. This review employed meta-analysis procedures to model VO 2max test precision.
ApproachStudies of the test-retest reliability of VO 2max protocols were identified from previous reviews and searches of computerized databases for biomedical, behavioral, and sports research. Of 51 studies identified, 12 were dropped because long test-retest intervals made it likely that true VO 2max values changed during the study. The reported means, standard deviations, and test-retest correlations were used to compute the standard error of measurement (SEM) for VO 2max for the remaining 39 studies. The age and gender composition of the sample were coded along with the exercise mode (treadmill, cycle ergometer, other) and the test-retest interval for the protocol. Meta-analysis produced a predictive model for SEM based on sample and protocol attributes.
ResultsAverage SEM was 2.58 ml·kg -1 ·min -1. SEM was higher in samples with higher average VO 2max . Age, gender, test interval, and exercise mode were not related to SEM. After allowing for outliers, the final model to predict SEM was ln(SEM) = 0.661 + (.006 * VO 2max ).
ConclusionsSEM increases as the average VO 2max of the sample increases. Other population and protocol attributes were not related to SEM. The potential applications of the model for SEM include evaluating new VO 2max protocols, evaluating field tests (e.g., run tests, walk tests), and making allowances for measurement error when investigating the relationships of VO 2max with other variables.