Affiliations: The University of Texas Medical Branch, Galveston, TX, USA
Correspondence:
[*]
Address for correspondence: Martha R. Hinman, EdD, PT, Associate Professor, Department of Physical Therapy, Rt. 1144, The University of Texas Medical Branch, 301 University Blvd., Galveston, TX 77555-1144, USA. Tel.: +1 409 772 9493; Fax: +1 409 747 1613; E-mail: mhinman@utmb.edu
Abstract: The purpose of this study was to establish the interrater reliability of the flexicurve when used by novice testers to measure thoracic kyphosis and lumbar lordosis in a community-based population. Three graduate students independently measured 51 healthy women (aged 21 to 88 years). The flexicurve was molded to the contour of the spine and traced onto a plain piece of paper to calculate the Index of Kyphosis (IK) and Index of Lordosis (IL). Subjects were measured while standing in their usual relaxed posture and again in their maximally erect posture. Intraclass correlation coefficients (ICCs) were calculated to determine the reliability among the three testers. IK measures were more reliable than IL measures with ICCs of 0.94 and 0.93 for the relaxed and erect postures, respectively. ICCs for the IL were 0.60 when relaxed and 0.73 when maximally erect. Greater variability in the lumbar measures may be attributed to difficulty in conforming the flexicurve to the smaller, concave curvature of the lumbar spine and interference from some subject's clothing. Nevertheless, these findings indicate good reliability for flexicurve measures of thoracic kyphosis and support the use of this measurement when documenting clinical changes in posture or identifying postural abnormalities in community-based screenings.
