An interview technique for recording work postures in epidemiological studies. Music-Norrtälje Study Group

The aim of the study was to present and evaluate a work-task-oriented interview technique focusing on the placement of the hands relative to the body and assessing per cent time spent in five standard work postures during a working day.

The reproducibility of estimated time spent in each work posture was tested by the test-retest method in 32 subjects; 16 were interviewed by the same interviewer and 16 were interviewed by another one at the retest. The validity concerning estimated time spent in th five standard work postures was tested in relation to observations in 58 male blue-collar workers. The mean registration (assessment) time was 6 hours and 15 minutes.

No evident differences in the reproducibility depending on same or different interviewers at test and retest could be observed. The linear relationship between times estimated by the interview and by observations was high for four of the work postures: 'sitting' (r = 0.86), 'standing with hands above shoulder level' (r = 0.87), 'between shoulder and knuckle level' (r = 0.75), and 'below knuckle level' (r = 0.93). When the work posture 'standing with hands between shoulder and knuckle level' was divided into 'hands fixed' (r = 0.62) and 'hands not fixed' (r = 0.50) the correlations were weak. Current musculoskeletal complaints did not influence the accuracy of the estimations.

The present task-oriented interview technique may be the best available method to estimate these work postures in a way that requires few resources compared to observations and technical measurements.