Objective. To investigate interobserver agreement between general practitioners and physiotherapists in diagnosing shoulder disorders. Design. Descriptive and comparative. Setting. Primary care practices in Amsterdam and Waterland, the Netherlands. Method. Eighteen general practitioners recruited 349 patients with a new episode of shoulder pain for an observational study. Interobserver agreement (Cohen's kappa) between general practitioner and physiotherapist (n = 13) was studied in a subset of 120 patients referred for physiotherapy. The patients were classified according to the diagnostic guidelines for shoulder pain issued by the Dutch College of General Practitioners in 1990. Results. The overall kappa for the classification of shoulder disorders was 0.31 (95% confidence limits: 0.19-0.42). Agreement was better in patients with capsular syndrome (0.48) and in patients with relatively long duration of symptoms (0.47). Conclusions. Interobserver agreement between general practitioners and physiotherapists was only fair. Many decisions concerning treatment are based on diagnosis, however. If a less detailed classification of shoulder complaints will suffice, less time and effort need to be spent on implementing a detailed diagnostic classification system in general practice.
|Translated title of the contribution||Diagnostics of shoulder complaints: Interobserver agreement between general practitioners and physiotherapists mediocre|
|Number of pages||4|
|Journal||Nederlands Tijdschrift voor Geneeskunde|
|Publication status||Published - 30 May 1998|