A method to characterize the coordination of body movements during a prolonged repetitive isodynamic lifting task was developed. Body movements, defined by angles at the hip and knee, were described by hyperbolic tangent models, with parameters of starting angle, ending angle, midpoint, and risetime calculated separately for each lift. The temporal midpoint and the risetime, normalized by the total duration of the lift, were used to describe the timing of the lift, and differences between the hip and knee midpoints and between the hip and knee risetimes described the relative timing, or coordination, of the hip and knee movements. The method was applied to data from a group of control subjects and a group of patients with chronic lower back pain. Control subjects tended to use a lifting pattern in which the two joints finished the movement at the same time, while patients more frequently used a movement pattern in which the two joints finished the lifting movement at different times. These differences suggest that the relative movement of the hip and knee may be important to consider when evaluating lifting differences between chronic low back pain patients and control subjects.