Photographic measurement of upper-body sitting posture of high school students: a reliability and validity study

All the reported measures of sitting posture, as well as photographs, have one flaw, as these measures are external to the body. These measures use calculations from external bony landmarks to estimate spinal posture, on the understanding that what is being measured externally reflects the shape, health and performance of structures of the underlying spine. Without a comparative measure of the relative position of the structures of the spine, the validity of any external spinal posture measure cannot be established. This paper reports on a study which tests the validity of photographs to measure adolescent sitting posture.

The study was conducted in a laboratory at the Department of Human Biology, University of Cape Town. A random sample of 40 adolescents were recruited from the Cape metropolitan schools, to detect differences of three degrees or more between the repeated measures of upright, normal or slouched posture (photographs) and between the posture photographs and LODOX measures. Eligible participants were healthy male and female subjects aged 15 or 16 years old, in Grade 10, and who were undertaking Computer or Computype studies at their schools. Two posture measurement tools were used in the study, namely: Photographs were taken using the Photographic Posture Analysis Method (PPAM) and Radiographs were taken using the LODOX (LODOX (Pty) Ltd) system. Subjects' posture was assessed in simulated computer workstations. The following angles were measured: the sagittal head angle, cervical angle, protraction/retraction angle, arm angle and the thoracic angle.

Data from 39 subjects (19 males, 20 females) was used for analysis (17 15-year-olds (7 boys and 10 girls), 22 16-year-olds (12 boys and 10 girls)). All but one photographic angle showed moderate to good correlation with the LODOX angles (Pearson r values 0.67-0.95) with the exception being the shoulder protraction/retraction angle Pearson r values. Bland Altman limits of agreement illustrated a slight bias for all angles. The reliability study findings from repeated photographs demonstrated moderate to good correlation of all angles (ICC values 0.78-0.99).

The findings of this study suggest that photographs provide valid and reliable indicators of the position of the underlying spine in sitting. Clinically it is important to know whether a patient is showing true progression in relation to a postural intervention. Based on the results of this study, the PPAM can be used in practice as a valid measure of sitting posture.