Abstract: The purpose of this study was to predict 3D cumulative L4/L5 spine loads and moments incurred during non-occupational tasks, from heart rate determined physical activity level (HR-PAL). Twelve subjects were videotaped while performing activities in their own homes. HR was continuously recorded during video collection and was subsequently used to calculate the PAL over the course of the 2-hour collection session. Simple regression revealed that between 76% and 82% of the variance in 3 of the 13 cumulative load measures studied (cumulative compression force and cumulative flexion and right axial twist moments) was accounted for by HR-PAL. Four additional cumulative output variables approached statistical significance. Cumulative compression force was the best predicted of all measures studied. Predicted and actual loads were not different from each other for all significant load measures. This initial study suggests that the use of heart rate for predicting cumulative compression shows potential as a simple method to track extended periods of cumulative exposure. Future work is planned to test this method in a number of industrial settings.