Reliability of static assessments of posture

Background and Purpose: There are several tools used for the assessment of posture, each with high validity and reliability. However, some of these tools are outdated, use crude methods or can be expensive. With the advances in photography and computer software, we can develop new ways to assess posture that are valid, reliable and easy to use. The purpose of this study was to examine the reliability of a new postural assessment method. Subjects and Methods: Fifty subjects from the California State University, Northridge student population participated in this study. Live and photographic measurements were performed to assess each subject's lumbar curvature. A goniometer was used to determine the amount of pelvic tilt in each individual, while the inclinometer was utilized to determine the angles found at the Ll and S1 vertebrae. Lumbar curvature (C) was obtained by adding the values at L1 and S1 (C = L1 +S1). Photographic analysis consisted of taking two separate photographs of each subject to assess the pelvic tilt and lumbar curvature. Reflective markers were attached to the ASIS, PSIS, L1, and S1 bony landmarks to improve the assessment technique. All photographs were uploaded to a desktop computer and measurements were taken directly from the monitor. Lumbar curvature was assessed in two ways: 1) using the radius (2-1 inches) of three circles of varying diameters ( 4-2 inches) 2) determining the inclination angles at L1 and S1. Lumbar curvature (C) was determined by adding the values of L1 and S1 (C = L1 +S1). Results-Live: The confidence interval (CI) indicated a difference in intrarater reliability between days for both testers. Inter-rater reliability was low to moderate, but did not differ between days. The ICC for the pelvic tilt measurement was low for both testers and no difference in intra- and inter-rater reliability. Results-Photoanalysis: Photographic inclination angles produced a difference in between day intra-rater reliability according the Cl. No difference was observed with inter-rater reliability between days. Pelvic tilt analysis indicated low to moderate intra-rater reliability, but no observed difference according to the Cl. Radius of the curvature results showed the highest intra- and inter-rater reliability. Discussion and Conclusions: Using the radius of the circle to determine lumbar curvature appears to be the most useful technique because reliability was found to be high among testers. In addition, this technique produced better inter-rater reliability. The use of computer software to assess lumbar curvature proved to require more training than anticipated as this may increase reliability of the technique. Utilizing an inclinometer to assess lumbar posture proved to be more reliable and easier to use than photographic analysis.