Article

Objectives: To determine if a classification of shoulder complaints in general practice can be made from variables of medical history and physical examination with nonmetric multidimensional scaling and to investigate the reproducibility of results from an earlier hierarchical cluster analysis. Design: A classification study performed using nonmetric multidimensional scaling. Setting: Four general practices in the Netherlands. Participants: Ninety-eight consecutive patients presenting with shoulder complaints in general practice were examined at study entry and after 2 weeks of treatment. Intervention: All patients were treated with a nonsteroidal anti-inflammatory drug during the first 2 weeks of treatment. Main Outcome Measures: Eleven variables of the medical history and 19 variables of the physical examination were used in a nonmetric multidimensional scaling analysis. Results: The analysis of the data at inclusion as well after 2 weeks shows that a 1-dimensional configuration can be used to represent the shoulder complaints. The results of the cluster analysis are consistent with the results of the nonmetric multidimensional scaling. The degree of limitation in range of motion and the degree of pain felt by the patients together determine the position of the patients on the dimension. Conclusions: The analysis shows that detailed classifications for the diagnosis of patients with shoulder complaints in general practice cannot be confirmed by empirical data available to the general practitioner. The results confirmed the recommendations of the new Dutch Guidelines for Shoulder Complaints, issued in 1999. (C) 2003 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation.

Langue : ANGLAIS

Identifiant basis : 2003227359