Rationale: Equations for predicting maximal exercise capacity are well accepted. An unexamined assumption is the accuracy of equations for predicting the submaximal aerobic cost of exercise training, which were developed in young/healthy individuals and are untested in older and/or clinical populations.

Objectives: This study was designed to test the hypothesis that the ACSM prediction equations could accurately predict Vo₂ during treadmill walking in a population of middle-aged and older adults.

Methods: Eleven middle-aged and older adults (age range = 55-85) free of limiting orthopedic issues, three were ambulatory without assistive device with a history of cerebrovascular accident (CVA) and 9 healthy young controls performed a maximal treadmill exercise test to determine peak Vo₂ and ventilatory threshold (VT). They returned > 48 hours later to perform submaximal treadmill walking consisting of four 6-minute stages at exercise intensities approximating 65%, 75%, 85%, and 95% of VT.

Results: The observed vs predicted Vo₂ in the middle-aged/older adults (9.6 vs 12.7 mL min^-1 kg^-1), was 95 +/- 12% vs the ACSM equation for speed/grade. In the young healthy controls (12.7 vs 14.6 mL min^-1 kg^-1) it was 87 +/- 10% of predicted, and in the 3 patients with a history of CVA (12.0 vs 10.0 mL min^-1 kg^-1) it was 119 +/- 13% of predicted.

Conclusions: The results support the experimental hypothesis that the ACSM treadmill walking equation for prediction of submaximal Vo₂ are accurate in middle-aged and older adults. However, this equation somewhat overpredicts Vo₂ in young individuals and under predicts Vo₂ in patients with a history of CVA.