Abstract
The statistical quality of 103 original articles published in The Journal of Bone and Joint Surgery (British Volume) in 1984 was assessed. Some papers were found to be deficient: thus, it was not always clear how series were selected and sometimes neither the data nor the results were clearly presented. Sample sizes were frequently inadequate for the conclusions reached and statistical techniques should have been used more frequently. A majority of papers were descriptions of case series for which no comparative data were made available. It is suggested that collaborative research would best advance knowledge about the relative benefits of various managements, and that statistical advice could make a substantial contribution.