The influence of the resistive load on the phase difference (PD) between chest wall and mouth flow in patients with chronic obstructive pulmonary disease (COPD) was investigated, and the factors that induce an enlargement of the PD were assessed. After a resistance tube (R = 2.8 hPa/l/s) was applied to the mouth of the subjects, the PDs increased significantly both in normal (pre: 2.20 +/- 0.89, post: 5.60 +/- 2.04, P < 0.01) and in COPD (pre: 10.86 +/- 3.81; post: 13.12 +/- 3.64, P < 0.01). A significant correlation between the airway resistance and PD was noted (r = 0.730, P < 0.001). These results suggest that airway resistance is an important determinant for the PD. The predicted phase difference (PD') was then calculated by the Runge-Kutta method, using the measured chest flow, the time domain functions of airway resistance and the thoracic gas volume during a respiratory cycle, assuming that the respiratory system was a single compartment. The PD/PD' at pre-resistive load was significantly larger in COPD than in normal subjects (normal: 1.03 +/- 0.11; COPD, 1.45 +/- 0.19, P < 0.01), whereas the PD/PD' at a post-resistive load was significantly smaller than at the pre-resistive load in COPD patients. These results agree well with the two parallel or serial compartment model with inhomogeneous airway resistance, where the mechanical property of the partitioning tissue between the two compartments was taken into account. The pattern of the increase in the work energy consumed within the airway lumen calculated from this model was almost the same as for PD. Thus, we conclude that the PD may be a good index of the overload of the respiratory movement and available to predict muscle fatigue in COPD patients.