gms | German Medical Science

Kongress Medizin und Gesellschaft 2007

17. bis 21.09.2007, Augsburg

Measures based on ICF categories comply with modern test theory and have discriminative and predictive validity

Meeting Abstract

Suche in Medline nach

  • Eva Grill - Institut für Gesundheits- und Rehabilitationswissenschaften, Ludwig-Maximilians-Universität München, München
  • Gerold Stucki - Klinik für Physikalische Medizin und Rehabilitation der Ludwig-Maximilians-Universität München, München

Kongress Medizin und Gesellschaft 2007. Augsburg, 17.-21.09.2007. Düsseldorf: German Medical Science GMS Publishing House; 2007. Doc07gmds817

Die elektronische Version dieses Artikels ist vollständig und ist verfügbar unter:

Veröffentlicht: 6. September 2007

© 2007 Grill et al.
Dieser Artikel ist ein Open Access-Artikel und steht unter den Creative Commons Lizenzbedingungen ( Er darf vervielfältigt, verbreitet und öffentlich zugänglich gemacht werden, vorausgesetzt dass Autor und Quelle genannt werden.



Introduction: The measurement of human functioning is one of the key aspects of research on effective patient care. Measures have to be comprehensive to catch all patient-relevant aspects and they have to be brief to be practical. Current measures, however, are not etiologically neutral, vary widely with respect to included items, cover but a restricted spectrum of functioning, are not invariant across different health conditions, and they are not adequate for parametric statistics. The International Classification of Functioning, Disability and Health (ICF) is a commonly accepted terminology for human functioning. The ICF can be used to document a categorical profile of a patient’s functional status. Nevertheless, up to now, there are no scales directly based on ICF categories. The objective of this study was to explore whether it is possible to develop scales based on sets of ICF categories, and, given these scales, if they have discriminative and predictive validity.

Methods: Patients with musculoskeletal conditions were recruited from 20 acute hospitals across Switzerland as part of a larger prospective cohort study. We collected information on the second level ICF categories with particular relevance for musculoskeletal conditions. We used the Rasch model which is part of Item Response Theory to create and validate scales.

Results: 142 patients (mean age 54, range 18 to 91 years, 49% female) were included in the analyses. We created two separate scales, one for upper extremity mobility, and one for lower extremity mobility. Model fit of both scales was acceptable, indicating unidimensionality. The two scales were able to discriminate between diagnostic groups, and were independent indicators of discharge destination.

Conclusion: It is possible to develop internally consistent ICF-based measures at interval level by applying ICF categories allowing for a more differentiated analysis to assess functioning.