Acta of Bioengineering and Biomechanics Vol. 4, No. 2, 2002 Predictive torque equations for joints of the extremities TADEUSZ BOBER Academy of Physical Education, Biomechanics Laboratory, ul. Paderewskiego 35, 51-612 Wrocław, Poland, tel. 347 3248, e-mail: tbober@awf.wroc.pl KORNELIA KULIG Department of Biokinesiology and Physical Therapy, University of Southern California, Los Angeles, CA JUDITH M. BURNFIELD Department of Biokinesiology and Physical Therapy, University of Southern California, Los Angeles, CA BOGDAN PIETRASZEWSKI Academy of Physical Education, Biomechanics Laboratory, ul. Paderewskiego, 35, 51-612, Wrocław, Poland The purpose of this study was to develop regression equations predicting torque output throughout the range of motion for the human elbow, shoulder, knee, and hip. Twenty-two healthy males participated. Torque values throughout the sagittal plane range of motion (i.e., flexion and extension) of the right elbow, shoulder, knee and hip were recorded (isokinetic dynamometer, 1 rad/sec) and expressed as a percentage of the peak torque produced for each motion. For each joint tested, regression equations based on ensemble- averaged, normalized torque data were calculated to predict the relative torque throughout the investigated range of motion when torque in one angular position was known. Shoulder flexion was best described by a second-order polynomial, while shoulder extension, elbow and knee flexion and extension, and hip flexion were described best by third-order polynomials. Hip extension was best described by a fourth-order polynomial. The regression multiple R 2 values ranged between 0.998 and 1.000. These regression equations can be used to predict the expected torque anywhere in the sagittal plane range of motion based on knowledge of a torque recorded at another joint angle for the elbow, shoulder, knee, or hip joint. These equations can be