Article Text

FRI0432 Comparison of tests for lumbar flexion and hip function in patients with axial spondyloarthritis and in healthy controls
  1. C. Stolwijk1,2,
  2. S. Ramiro3,4,
  3. R. Landewé3,5,
  4. D. van der Heijde6,
  5. A. van Tubergen1,2
  1. 1Rheumatology, Maastricht University Medical Center
  2. 2School for Public Health and Primary Care (CAPHRI), Maastricht University, Maastricht
  3. 3Rheumatology, Academic Medical Center, Amsterdam, Netherlands
  4. 4Rheumatology, Hospital Garcia de Orta, Almada, Portugal
  5. 5Rheumatology, Atrium Medical Center, Heerlen
  6. 6Rheumatology, Leiden University Medical Center, Leiden, Netherlands


Background Different approaches exist to assess limitation in spinal mobility and hip function in axial spondyloarthritis (axSpA). The 15cm-Schober’s test (originally included in the Bath AS mobility index (BASMI)), and the 10cm-Schober’s test (recommended by the ASAS to use in the BASMI), are both used to quantify lumbar flexion. The intermalleolar distance (IMD) is included in the BASMI for measuring hip function. A substitute for the IMD, the internal hip rotation (IHR) was later proposed, and is part of the Edmonton AS Metrology Index (EDASMI) [1]. However, direct comparisons between the different approaches are lacking.

Objectives To assess the agreement between both the 10cm- and 15cm- Schober’s test and the IMD and the IHR, and their performance in composite indices.

Methods A cross-sectional study was conducted among both healthy volunteers (HVs) aged 20-69 years old (‘Mobility study’) stratified for gender, age- and height-categories, and consecutive outpatients with axSpA. The spinal and hip mobility tests were assessed in all subjects. Agreement between the 10cm- and 15cm-Schober’s test and between the IMD and IHR were assessed by Pearson correlation coefficients, visually by cumulative probability plots, and by Bland-Altman plots. The performance of the two versions of the Schober’s test in BASMI10 was assessed by calculating the BASMI-score using either method.

Results A total of 393 HVs (50.9% males; mean age 43.9 years) and 77 patients (50.6% males; mean age 47.9 years; 76.6% fulfilled the modified New-York criteria) were included. The 10cm- and 15m-Schober’s test had a mean value of 5.0cm (SD 1.0) and 6.3cm (SD 1.2) in HVs, and 3.7cm (SD 1.5) and 5.2cm (SD 1.9) in patients, respectively. A strong correlation between the 10cm- and 15cm-Schober’s test was found, both in HVs (r = 0.89) and in patients (r = 0.92). The Bland-Altman plot showed a systematic difference between the 15cm- and 10cm-Schober’s test of 1.3cm (SD 0.5) in HVs and 1.5cm (SD 0.8) in patients. This difference resulted in an increased mean total BASMI10 score of 0.3 points in HVs and 0.4 points in patients with the 15cm-Schober compared with the 10cm-Schober. The IMD had a mean value of 112cm (SD 14) in HVs and 100cm (SD 20) in patients. The IHR had a mean value of 48cm (SD 10) in HVs and 45cm (SD 10) in patients. The correlation between the IMD and IHR was weak both in HVs (r = 0.29) and in patients (r = 0.37). A wide distribution of scores and no tendency for parallel increases between the IMD and IHR was found.

Conclusions The 10cm- and 15cm-Schober’s test are highly correlated but the 15cm-method results in systematically higher scores than the 10cm-method, which influences the total BASMI score. Therefore, these methods cannot be used interchangeably. The IMD and IHR are not correlated; the IHR cannot be used as a substitute for the IMD.


  • Maksymowych et al. A&R 2006;55:575-82

Disclosure of Interest None Declared

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.