A Comparison Of Step Tests To Predict Maximal Oxygen Consumption

Abstract

INTRODUCTION: Of the many methods to estimate maximal oxygen consumption (VO2max), step tests provide a quick estimation of VO2max without expensive equipment. Few studies have investigated varied step height and cadence speed across a broad sample population. The purpose of this study was to generate and validate models to predict VO2max using three different step tests. METHODS: A sample of 162 adults, aged 18-79 years, completed a graded exercise test to assess VO2max. All three step tests were completed on a later day. The step tests differed in duration, number of stages, and step height. Test 1 (T1) was a two-stage, six-minute test, and had a six-inch step height, Test 2 (T2) was a three-stage, nine-minute test, and had a six-inch step height, and Test 3 (T3) was a three-stage, nine-minute test, and had an eight-inch step height. Stepping cadence increased with each stage and was prescribed based on age group (<40 years, 40-60 years, and >60 years). Recovery heart rate (HRrec) was measured every 30-seconds for two-minutes after each test. Model equations were obtained using hierarchal multiple regression analysis. Covariates for each model included age, sex, body mass, stepping cadence, and HRrec at all measured time intervals. The baseline model for each test consisted of age, sex, and body mass. Stepping cadence was entered in step two but was only significant for T3. Thirty-second HRrec was entered in the final step and was significant for all three tests. Validity was assessed using a Jackknife cross-validation method, allowing for root mean square error (RMSE) and percent bias to be calculated. RESULTS: The baseline model accounted for ~72% of the variance for predicting VO2max. Each model accounted for ~83% of VO2max variance and had an error ~4.15 ml.kg-1.min-1. The results of the jackknife analysis found that bias for all three tests was minimal (<0.001%) and the resulting bias-adjusted R2 was 0.825, 0.826, and 0.834 for T1, T2, and T3, respectively. Adjusted RMSE was 4.259, 4.223, and 4.102 ml.kg-1.min-1 for T1, T2, and T3, respectively. DISCUSSION: All three step tests account for ~83% of the variance of VO2max with an error around 4.2 ml.kg-1.min-1. As there is a minimal difference between tests, there is flexibility in the utilization of these step tests based on the need of the administrator and ability of the client.