Email updates

Keep up to date with the latest news and content from BMC Musculoskeletal Disorders and BioMed Central.

Open Access Research article

Reliability of upright posture measurements in primary school children

Maureen P McEvoy1* and Karen Grimmer12

Author Affiliations

1 School of Health Sciences, University of South Australia, North Tce., Adelaide, 5000, Australia

2 Centre of Allied Health Evidence, collaborating centre of the Joanna Briggs Institute, GPO Box 2471, Division of Health Sciences, University of South Australia, North Tce., Adelaide, 5000, Australia

For all author emails, please log on.

BMC Musculoskeletal Disorders 2005, 6:35  doi:10.1186/1471-2474-6-35

Published: 29 June 2005



Correct upright posture is considered to be a measure of good musculoskeletal health. Little is known about the usual variability of children's upright standing posture. The aim of this study was to assess differences between repeated measures of upright posture in a group of primary school children.


Sagittal plane photographs of usual, relaxed upright standing posture of 38 boys and girls aged 5–12 years were taken twice within an hour. Reflective markers were placed over the canthus, tragus, C7 spinous process, greater trochanter and lateral malleolus. Digitising software was used to calculate the x,y plane coordinates, from which five postural angles were calculated (trunk, neck, gaze, head on neck, lower limb). Height, weight, motor control estimates (as measured by the Brace Tests) and presence of recent pain were recorded for each child, and the association between the first test measure of posture angles and these factors was assessed using linear regression and ANOVA models. Multiple ANOVA models were applied to analyse the effect of repeated testing, and significant predictors on the angles.


Four of the five postural angles (trunk, neck, head on neck, lower limb) were significantly influenced by age. As age was strongly associated with height (r2 = 0.84) and moderately associated with weight and motor control (r2 = 0.67, 0.56 respectively), these developmental parameters may well explain the age effect on angles. There was no relationship between age and pain reported on either the testing day, or recently, and there was no gender influence on any angle. There was no significant effect of repeated testing on any angle (ICC>0.93). None of the hypothesized predictors were associated with differences in angles from repeated testing.


This study outlined the variability of relaxed upright standing posture of children aged 5–12 years, when measured twice in an hour. Age influenced the size of the angles but not the variability. While the subject numbers in this study are small, the findings provide useful information on which further studies in posture and its development in pre-adolescent children can be based.