To evaluate whether individualized ramp protocols may be better than step protocols in patients > or = 60 years of age referred for exercise testing, peak cardiopulmonary responses and accuracy in prediction of oxygen uptake (VO2) for individualized ramp and step protocols (Bruce or modified Bruce) were compared. Twenty-four subjects (67+/-3 years) with known or suspected coronary artery disease performed both tests in random order. Protocols were selected based on estimated exercise capacity using a pretest activity questionnaire. No differences were observed between peak VO2 (19.3+/-6.3 and 19.1+/-6.4 ml/kg/min), heart rate (127+/-15 and 126+/-16 beats/min), rate-pressure product (24.0+/-4.8 and 23.4+/-4.9 beats/min x mm Hg x 10(3)) and anaerobic threshold (16.6+/-3.7 and 16.0+/-4.7 ml/kg/min) for the ramp and step protocols, respectively. The relation between measured submaximal VO2 and American College of Sports Medicine (ACSM)-predicted VO2 during the ramp protocol is demonstrated by the regression coefficient (beta), where beta = 0.92 (95% confidence intervals [CI] 0.85 to 0.99) and for the step protocols where beta = 1.02 (95% CI 0.84 to 1.20). Peak cardiopulmonary responses in the elderly are similar during individualized ramp and step protocols when appropriately selected based on a pretest activity questionnaire. Both protocols appear to provide clinically reasonable estimates of VO2 when gas exchange analysis is not used.