Effects of pain on shoulder jerk in elderly subjects during repetitive lifting

The purpose of this study was to determine if jerk, (rate of change of acceleration) differs in elderly control and CLBP patients during a lifting task. Jerk was calculated for the shoulder trajectory, since shoulder motion depends on motion at both the hips and knees. To reduce noise, Woltring´s GCV hepatic spline was fitted to the data and derivatives were calculated through differentiation of the spline coefficients. A root mean square (rms) measure of jerk was used for comparison. The rms jerk values for the lifts were designated as early, middle or late based on the number of repetitions completed by the subject, and a repeated measures analysis of variance allowed testing for changes over time during the task. Lift duration was also analyzed. Significant group differences were found for lift duration and rms jerk, with controls lifting faster and generating greater values of jerk than CLBP patients. Changes over time were also observed in which lift duration decreased and jerk increased with task time. A group by time interaction occurred for lift duration but not for jerk. In general, patients had slower lifts with lower jerk values than controls, suggesting that pain impacts lifting style.