rss
Ann Rheum Dis 2008;67:375-379 doi:10.1136/ard.2007.072785
  • Extended report

Reliability and sensitivity to change of the Simple Erosion Narrowing Score compared with the Sharp–van der Heijde method for scoring radiographs in rheumatoid arthritis

  1. E M Dias1,
  2. C Lukas1,
  3. R Landewé1,
  4. S Fatenejad2,
  5. D van der Heijde3
  1. 1
    Department of Internal Medicine, Division of Rheumatology, University Hospital Maastricht, Maastricht, the Netherlands
  2. 2
    Wyeth Research, Collegeville, PA, USA
  3. 3
    Department of Rheumatology, Leiden University Medical Center, Leiden, the Netherlands
  1. Dr Robert Landewé, Department of Internal Medicine, Division of Rheumatology, PO Box 5800, 6202 AZ Maastricht, The Netherlands; rlan{at}sint.azm.nl
  • Accepted 17 July 2007
  • Published Online First 20 July 2007

Abstract

Objective: To compare the performance of a simplified scoring method for structural damage on radiographs of patients with rheumatoid arthritis (the Simple Erosion Narrowing Score or SENS) with the Sharp–van der Heijde Score (SHS) as reference.

Method: We used the radiographic data from the Trial of Etanercept and Methotrexate with Radiographic Patient Outcomes (TEMPO trial). The SENS was derived from the crude SHS data. Inter-observer reliability for status scores and change scores was determined by intraclass correlation coefficients and by the Smallest Detectable Change method. The ability to discriminate between treatment groups was assessed by the Mann–Whitney U test. Stratifying the sensitivity to change and discriminative ability for different levels of disease severity assessed a potential ceiling effect.

Results: Inter-observer reliability was similar for both methods. Intraclass correlation coefficients were higher for status scores than for change scores. The Smallest Detectable Change was 4.98 (1.1% of possible maximum score) for SHS and 2.28 (3.5%) for SENS. Sensitivity of SENS to detect progression above the Smallest Detectable Change, with reference SHS, ranged from 45.0 to 88.7%. Specificity ranged from 81.5 to 97.3%, and the κ coefficient (between-method agreement) ranged from 0.58 to 0.66. Discriminative ability between treatment groups was good and similar for both methods. A ceiling effect could not be detected.

Conclusions: With regard to most of the tested properties, the performance of SENS is as good as that of SHS. This confirms that SENS is a valuable method, which may be feasible in clinical practice.

Footnotes

  • Competing interests: None.

This Article

  1. Abstract
  2. Full text
  3. PDF
  4. All versions of this article:
    1. ard.2007.072785v1
    2. 67/3/375 most recent

Responses

  1. Submit a response
  2. No responses published

Register for free content

The full back archive is now available for all BMJ Journals. Institutional subscribers may access the entire archive as part of their subscription. Personal subscribers will also have access to all content when logged in. Non-subscribers who register have free access to all articles published before 2006 right back to volume 1 issue 1. Register here to access the free archive of all BMJ Journals.

Don't forget to sign up for content alerts so you keep up to date with all the articles as they are published.