Trunk posture: Reliability, accuracy, and risk estimates for low back pain from a video based assessment method

W. P. Neumann*, R. P. Wells, R. W. Norman, M. S. Kerr, J. Frank, H. S. Shannon

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


It has been recently reported that both dynamic movement characteristics, as well as the duration of postures adopted during work, are important in the development of low back pain (LBP). This paper presents a video-based posture assessment method capable of measuring trunk angles and angular velocities in industrial workplaces. The inter-observer reliability, system accuracy, and the relationship of the measured exposures to the reporting of low back pain are reported. The video analysis workstation consisted of a desktop computer equipped with digital video capture and playback technology, a VCR, and a computer game type joystick. The operator could then use a joystick to track trunk flexion and lateral bending during computer-controlled video playback. The joystick buttons were used for binary input of twisting. The inter-observer reliability for peak flexion and percentage of time spent in posture category variables were excellent (ICC>0.8). Lower reliability levels were observed for peak and average velocity and movement related variables. The video analysis system time series data showed very high correlation to the criterion optoelectronic imaging system (r = 0.92). Root mean square errors averaged 5.8° for the amplitude probability distribution function data. Trunk flexion variables including peak level, peak velocity, average velocity indicators, and percent time in flexion category indicators all showed significant differences between cases and controls in the epidemiological study. A model consisting of the measures peak trunk flexion, percent time in lateral bend and average lateral bending velocity emerged after multivariable analysis for relationship to low back pain. Relevance to industry: Risk of injury for the low back is multifactorial. The trunk position and movement velocity are emerging as important parameters. This analysis confirms the importance of these factors and demonstrates the utility of a video-based method to measure them in industrial settings.

Original languageEnglish
Pages (from-to)355-365
Number of pages11
JournalInternational Journal of Industrial Ergonomics
Issue number6
Publication statusPublished - 17 Oct 2001


  • Accuracy
  • Epidemiology
  • Kinematics
  • Low back pain
  • Posture
  • Reliability

Fingerprint Dive into the research topics of 'Trunk posture: Reliability, accuracy, and risk estimates for low back pain from a video based assessment method'. Together they form a unique fingerprint.

Cite this