Reliability and validity of measures taken during the Chester step test to predict aerobic power and to prescribe aerobic exercise

To evaluate the reliability and validity of measures taken during the Chester step test (CST) used to predict VO(2)max and prescribe subsequent exercise.

The CST was performed twice on separate days by 7 males and 6 females aged 22.4 (SD 4.6) years. Heart rate (HR), ratings of perceived exertion (RPE), and oxygen uptake (VO(2)) were measured at each stage of the CST.

RPE, HR, and actual VO(2) were the same at each stage for both trials but each of these measures was significantly different between CST stages (p<0.0005). Intertrial bias +/-95% limits of agreement (95% LoA) of HR reached acceptable limits at CST stage IV (-2+/-10 beats/min) and for RPE at stages III (0.2+/-1.4) and IV (0.5+/-1.9). Age estimated HRmax significantly overestimated actual HRmax of 5 beats/min (p = 0.016) and the 95% LoA showed that this error could range from an underestimation of 17 beats/min to an overestimation of 7 beats/min. Estimated versus actual VO(2) at each CST stage during both trials showed errors ranging between 11% and 19%. Trial 1 underestimated actual VO(2)max by 2.8 ml/kg/min (p = 0.006) and trial 2 by 1.6 ml/kg/min (not significant). The intertrial agreement in predicted VO(2)max was relatively narrow with a bias +/-95% LoA of -0.8+/-3.7 ml/kg/min. The RPE and %HRmax (actual) correlation improved with a second trial. At all CST stages in trial 2 RPE:%HRmax coefficients were significant with the highest correlations at CST stages III (r = 0.78) and IV (r = 0.84).

CST VO(2)max prediction validity is questioned but the CST is reliable on a test-retest basis. VO(2)max prediction error is due more to VO(2) estimation error at each CST stage compared with error in age estimated HRmax. The HR/RPE relation at >50% VO(2)max reliably represents the recommended intensity for developing cardiorespiratory fitness, but only when a practice trial of the CST is first performed.