Study of validity and intra and inter-observer reliability of modified-modified Schöber test in subjects with low-back pain

Abstract

In patients with low-back pain the lumbar spine range of motion (ROM) is often measured by the modified version of the modified Schöber test (MMST), but its psychometric properties have not been ascertained for clinical use. The purpose here was to verify intra and inter-observer validity and reliability of the MMST in subjects with low-back pain, and to compare obtained ROM measures to those obtained by radiography, taken as gold standard. The study involved 20 subjects with chronic low-back pain, of both sexes, employees at a university hospital. The MMST was applied twice by two examiners each. The Pearson correlation coefficient found when comparing measures obtained via MMST and radiography was r=0.14, showing a poor correlation between the tests. The intra-observer intraclass correlation coefficient (ICC) found was 0.96 (CI 95% 0.91;0.98), and the inter-observer ICC was 0.93 (IC 95% 0.84;0.97), showing high reliability; the Bland & Altman agreement test showed high agreement intra (-0.21) and inter-observer (-0.21). Although a high reliability both intra and inter-observer was found for the modified-modified Schöber test, the latter showed low validity in assessing lumbar spine range of motion, when compared to the gold standard.