Indices of ventilatory efficiency have proven useful in assessing patients with heart and lung disease. One of these indices is the slope of the ventilation (V̇E) versus carbon dioxide output (V̇CO2) relationship during cardiopulmonary exercise testing (CPET) for work rates where the relationship is linear. However, this relationship is defined not only by the slope but also by the y-intercept. To examine whether this relationship is dependent on the speed of the CPET protocol, 30 healthy subjects (16 males) were administered a rapid CPET with 1-min increment duration (1-min CPET) to the limit of tolerance and a slow CPET with 4-min increment duration (4-min CPET) to the lactate threshold. Ventilation and the gas fractions for oxygen and CO2 were measured with a Vacumed metabolic cart. The average increment size of both protocols for both sexes was not significantly different (P>0·05). For the males, the mean (SD) slope for the 1- and 4-min CPET was 20·12 (2·61) and 20·37 (2·41), respectively. The corresponding values for the y-intercept were 4.·89 (2·08) and 5.·10 (2·00) l min−1. For the females, the mean (SD) slope for the 1- and 4-min CPET was 23·90 (2·38) and 24·16 (2·55), respectively. The corresponding values for the y-intercept were 3·93 (0·39) and 3·77 (0·71) l min−1. Paired t-test analysis demonstrated for both sexes that the slopes and y-intercepts were not different for the two protocols (P>0·05). The results of this study demonstrate that the V̇E versus V̇CO2 relationship is not dependent on the speed of the CPET protocol.