Cross-Validation of a PACER Prediction Equation for Assessing Aerobic Capacity in Hungarian Youth.

PURPOSE: The purpose of this article was to evaluate the validity of the Progressive Aerobic Cardiovascular and Endurance Run (PACER) test in a sample of Hungarian youth.
METHOD: Approximately 500 participants (aged 10-18 years old) were randomly selected across Hungary to complete both laboratory (maximal treadmill protocol) and field assessments (PACER) of aerobic capacity. Agreement between lab- and PACER-derived peak oxygen consumption (VO2) was examined using linear regression and 2-sided equivalence testing techniques, respectively. The impact of agreement on the classification accuracy of peak VO2 estimates into FITNESSGRAM® fitness zones was determined with kappa statistics.
RESULTS: The final sample resulted in a total of 167 boys and 143 girls (N = 310). Analyses revealed that lab and PACER VO2 shared 13% to 18% of their variance (R(2)boys = .13, R(2)girls = .18) and that limits of agreement ranged from - 39.9 mL/kg/min to +37.6 mL/kg/min depending on the sex. The absolute error values were 14% for boys and 16% for girls; however, the average peak VO2 estimates from the PACER were within the 10% equivalence region for girls (37.2 mL/kg/min to 45.4 mL/kg/min), but not for boys (45.2 mL/kg/min to 55.2 mL/kg/min). When lab and PACER VO2 were categorized according to Fitnessgram zones, agreement was fair for both sexes (boys, Kappa = .25, and girls, Kappa = .31).
CONCLUSIONS: Although the correlations between measured and predicted peak VO2 were lower than expected, the magnitude of error observed in the PACER is similar to past ranges of error observed in other studies (10%-15%), but the large individual error should be considered when interpreting individual results from this field-based measure of aerobic capacity. There was reasonable classification agreement between lab estimates and the PACER test for classification into the various fitness standards.